ACKNOWLEDGEMENTS

     The author would like express sincere gratitude to all who have supported, encouraged, facilitated, and supervised this dissertation. Appreciation is expressed to the husband, David R. Howard, and family of
the author for their support, especially to the author's parents,Colonel
and Mrs. H. S. Lowe, for the fine educational opportunities they provided the author.
     Special thanks is given to the author's committee for their excellent instruction, facilitation, and supervision. Thanks is expressed
to Dr. Jo Ann Sweeney and Pattie Rose for Guidance. Heartfelt gratitude is given to the author's mentors and role models: Claire Nace, Dr. Lucia McKay, Dr. Cecilia Erlund, and Dr. Linda Pehl. Appreciation is extended especially to Betty Sue Beebe, Berneta Peeples,dr. Rachel LaRoe, University of Mary Hardin-Baylor's Class of 1945, and to the others who participated in surveys, interviews, and oral histories in the author's research related to the female history of U.M.H.B.
     Thanks is expressed to the author's faculty, library, and media colleagues and staff at the U.M.H.B. who encouraged and assisted the doctoral studies and research of the author, especially to the following colleagues: Max Hart, Dr. Bob Holland, Dr. David Roland, Dr. George Stansbury, and Dr. Joseph Tabarlet. Special gratitude is expressed
Dr. Lucinda Harman for her expert assistance to the author in analyzing and graphing the data.  

ABSTRACT

     This study investigates locus of control as a changeable variable in first-year, female, coeducational, college students at the University of Mary Hardin-Baylor. Further, it investigates the relationship between locus of control in these subjects and their identification, or not, of female, faculty or staff, role models.Measurement instruments were Rotter's Internal-External Locus of Control Scale and the investigator's role model questionnaire. The study was built on historical research of the university's female-only days and on a review of literature in the fields of education, psychology, sociology, industry, labor, and business.

     This study presents evidence that suggests that the subjects who identified role models had stronger internal locus of control than the subjects who did not identify role models, that locus of control was a changeable variable, that internal locus of control increased over the first year of college in the role model group (positive effect), and that internal locus of decreased over the first year of college in the no role model group (negative effect). Thus, this study suggests a factor which may strengthen internal locus of control in first-year, female, coeducational college students.

                            TABLE OF CONTENTS

1. INTRODUCTION

   STATEMENT OF THE PROBLEM

   DEFINITION OF TERMS

   SIGNIFICANCE OF THE STUDY

   EMERGENCE OF RESEARCH QUESTIONS

   ORGANIZATION OF THE RESEARCH PROJECT

2. REVIEW OF THE LITERATURE

   EDUCATIONAL LITERATURE

   PSYCHOLOGICAL LITERATURE

   SOCIOLOGICAL LITERATURE

   INDUSTRIAL AND LABOR LITERATURE

3. RESEARCH DESIGN AND PROCEDURE

   RESEARCH QUESTIONS

   RESEARCH DESIGN

   HYPOTHESES

   SUBJECTS

   PROCEDURES

4. DATA ANALYSIS

   CONCLUSIONS

5. LIMITATIONS AND RESEARCH RECOMMENDATIONS

   LIMITATIONS

   FUTURE RESEARCH RECOMMENDATIONS

APPENDICES

Research Tools

Samples of Anecdotal Evidence of Female Role Models and Internal Locus of Control in Female Students at The University of Mary Hardin-Baylor During Its Female-Only Days

REFERENCES

INTRODUCTION

     Twentieth-century learning theories can be divided into two major categories: cognitivism and behaviorism.Cognitive theories, which recognize complex internal cognitive and social factors in students as significant in producing achievement motivation, have been challenged by behavioristic theories, which more simply and mechanically recognize external stimuli as factors in producing achievement motivation in students. (Lefrancois, 1988) " . . . traditional theories have painted an overly passive and mechanistic picture of human beings. The newer approaches are clearly cognitive." (Lefrancois, 1988: 278) In recent decades, cognitive learning theories have emerged which focus on achievement motivation in terms of student attributions and observations.

     Rotter provides a foundation for the cognitive theories of the last twenty years in his social learning theory, a social-personality theory that describes students in terms of their tendencies to attribute success or failure to internal or external factors.

     . . . the extent to which a person believes that he can control what
     happens to him is referred to as a belief in internal control of
     reinforcement. A belief that one is controlled by luck, fate, or     
     powerful others, is referred to as a belief in external control of
     reinforcement. (Rotter, 1971:61)

     R. de Charms uses the attributional terms "origin" and "pawn" in a similar way to Rotter's attributional terms "internal and external." (de Charms, 1968) "An Origin is a person who perceives his behavior as determined by his own choosing: a Pawn is a person who perceives his behaviour as determined by external forces beyond his control." (de Charms, 1968: 273-274)

     Weiner, likewise, presents a systematic achievement motivation theory which begins with the assumption that students attribute their successes and failures to internal or external causes. Those who attribute their successes to internal causes have a greater degree of achievement motivation than those who attribute their successes to external causes. Those who attribute their failures to external causes, rather than to internal causes, have stronger self-esteem and achievement motivation than those who attribute failure to internal causes. (Weiner: 1979; Teglasi 1978)

     One area of research that developed from Rotter's theory concerns the locus of control in individuals. The locus of control in people affects their perception of rewards they receive as controlled by internal factors, such as their personal ability, effort, or skill, or as controlled by external factors, such as chance,luck, or other people. Persons' perceptions of their environments as skill-or-chance-determined influence positive or negative shifts of achievement expectancy in themselves following success and failure. In addition, there are other characteristics that emerge as a result of individuals' perceptions of their environments as personally or externally controlled. For example, research investigations have suggested that people classified as internal in locus of control display more information-seeking behavior and make better use of this information than people classified as external in locus of control. (Weiner, 1979: 272)

     Subsequent investigations of "internality-externality" have established that it is a useful and valid way of classifying students in terms of how they attribute the rewards and reinforcements they receive following their actions and behaviors. (Dweck,1986) Such investigations may be especially important for females. Evidence suggests that, in American schools, males have the tendency to make more external attributions for failure than females. "American girls are more likely to blame themselves, attributing failure to lack of intelligence or ability. When the boy does well, however, he will often attribute his success to a personal trait." (Cushner, McClelland, and Safford, 1992: 32).

     Furthermore, educational theorists and researchers have been concerned in recent decades with females' "hypothesized fear of success." (Weiner, 1979: 182; Teglasi, 1978; Betz & Fitzgerald, 1987) Betz & Fitzgerald credit the original concept of "fear of success" to Horner and define it as the motive to avoid success, which in women is ". . . characterized by a stable dispositional tendency to become anxious about achieving success, which tendency is thought to reduce achievement motivation and inhibit achievement-related behavior." (Betz & Fitzgerald, 1987: 153)

     Finally, cognitive and social learning theories include the issue of learning by observation. Maccoby and Jacklin state that ". . . although various theorists of child development (such as Kagan, Kohlberg, Mischel, Mussen, & Sears) have differing theoretical viewpoints, all stress the influence of imitation and identification." (Basow & Howe, 1980: 559; Maccoby & Jacklin, 1974) Bandura asserts "Most human behavior is learned observationally through modeling: from observing others one forms an idea of how new behaviors are performed . . . . Models who possess engaging qualities are sought out . . . " (Bandura, 1977: 22-23) Muss presents a theory of adolescence in which he contends that a student's observation of her teacher may have a more powerful impact on the student than instructional methods. (Muss, 1975)

      Educational theorists, as well as being concerned about the issue of locus of control and its effect on achievement, especially in females, have been concerned about the issue of role modeling and its effect on achievement for females. Studies of students and teachers in the United States, Canada, England, and Nigeria present evidence of a positive relationship between same-sex role modeling and academic achievement. (Johnson, 1973-74) Studies provide evidence that same-sex role modeling may be especially important for female students. (Finn, 1980) Block asserts that proponents of single-sex schooling emphasize the importance of female faculty members as successful, competent role models and as sources of encouragement for female students. (Colangelo, Dustin, & Foxley, 1979; Block, 1979)

     Maccoby and Jacklin contend that adolescent female students display more concern about being liked and respected by their teachers, are more dependent on teachers for emotional support, and are more subject to suggestibility than male students. (Maccoby & Jacklin, 1974; Colangelo, Dustin, & Foxley, 1979) Block contends that the "effects of single-sex education may not be parallel for the two sexes . . . ." (Colangelo,
Dustin, & Foxley, 1979: 57, Block, 1979) Females may, in general, value relationships and cooperation more than males. Gilligan traces female psychological and moral development that places relationships and caring
for others at the center of the developing female self. (Gilligan, 1982)

     Belenky, Clinchy, Goldberger, and Tarule present an epistemological model of development based on the importance of relationships in cognitive development for females. Using interviews with 135 female students from a wide range of educational institutions, they conclude that females have a "connected way of knowing" based on experiential and relational modes of thought. Belenky et al. contend that females are especially influenced by relationships and are positively influenced by professional female role models. (Belenky et al., 1986)

     Studies present evidence that same-sex role modeling is especially important for female students and that female role models strengthen achievement in female college students. (Douvan, 1976; Lentz, 1980;
Tidball, 1980; Basow & Howe, 1980; Solnick, 1995; Canes & Rosen,1995;
Ehrenberg, 1995) Studies further support the idea that females can strengthen internal locus of control in other females. (Teglasi, 1978)

 STATEMENT OF THE PROBLEM

     Do female role models provide a solution to the problem of female student under-achievement and under-representation? Studies present evidence that American female students are outperformed and overrepresented by American male students in the fields of mathematics, science, and engineering. (Rossi, 1965; Ernest, 1976; Fennema & Sherman, 1977; McKay, 1978; Betz & Fitzgerald,1980; Ehrenberg, 1995) Ehrenberg asserts that a solution to the problem of poorer scores on standarized tests in science and mathematics for female students and to the problem of a lower percentage of females as students and faculty in mathematics, science, and engineering is to provide female faculty role models in these fields. (Ehrenberg, 1995)

     Educational institutions at all levels are also aggressively seeking
     to expand . . . female faculty to employ in 'traditionally male
     fields,' such as mathematics, science, and engineering. Females are
     under-represented in many of these fields at the collegiate level,
     both as students and as faculty. (Ehrenberg, 1995: 482; Ehrenberg,
     1992)

     Ehrenberg contends that a major reason for this under-representation of females in fields traditionally dominated by males is that by the time young women reach the end of high school, they tend to perform more poorly than males in mathematics and science in standarized tests. Ehrenberg asserts that

     . . . . Many point to the absence of female role models in science
     and mathematics . . . as part of the explanation for these outcomes . .
     . . The scarcity of female faculty in traditionally male fields in
     higher education is also believed to contribute to the reluctance of
     women to major in traditionally male fields of study, to their high
     propensity to switch to other majors if they start out in a   
     traditionally male field, and to the small number of women pursuing       doctoral study in many . . . male fields. (Ehrenberg, 1995: 482)

     Literature in the fields of education, psychology, sociology, science, labor, industry, and business addresses the following questions: Is there
a relationship between female role models and female college students' attitudes toward themselves and toward professional women? (Walker, 1984)
Is there a relationship between career salience in female college students and female role models? (Rossi, 1965; ; Douvan, 1976; Lentz, 1980; Tidball, 1980) Is female role modeling a significant factor in female students' achievement motivation? (Rossi, 1965; Douvan, 1976; Lentz, 1980; Tidball, 1980; Smith & Erb, 1986; Anderson & Ramsey, 1990; Solnick, 1995) Is there
a relationship between female achievement and affirmative action? (Tidball, 1973, 1980; Berger, 1990; Canes & Rosen, 1995) What are the barriers to female achievement? (Rossi, 1965; Betz & Fitgerald, 1987) Is there a relationship between females' attitudes and female role models? (McLure
& Piel, 1978; Basow & Howe, 1980; Walker, 1984). Do female role models
affect female earnings? (Angle & Wiseman, 1981; Gerhart, 1990; Fuller
& Schoenberger, 1991; Rothstein, 1995)

     Additional questions addressed include these: Is there a relationship between internal locus of control and persistence? (Strain, 1993) Is there
a relationship between internal locus of control and achievement? (Lewin
& Stephens, 1994) Is internal locus of control a significant factor in females' achievement motivation? (Teglasi, 1978) Is there a relationship between locus of control and motivation? (Stone & Jackson, 1975) Is there
a relationship between locus of control and modeling? (Stone & Jackson, 1975) Is there a relationship between locus of control in females and female competitors? (Teglasi, 1978) Is there a relationship between locus of control and cognitive ability? (Fry, 1975) Is there a relationship between change in observers and role models? (Schwartz & Hawkins, 1965; Myrick,1969; Marlatt, Jacobson, Johnson, & Morrice, 1970; Duke & Frankel, 1971; Sarason, Ganzer, & Singer, 1972; Stone & Stebbins, 1975; Stone & Gotlib, 1975; Dowling & Franz, 1975)

     Researchers have studied these questions and have provided evidence that suggests the following: Females who have positive attitudes toward professional women and who have positive self-esteem have female role models. (Walker, 1984) A positive relationship exists between career salience in female college students and female role models. (Rossi, 1965; Douvan, 1976; Lentz, 1980; Tidball, 1980; Walker, 1984) Female college students who have female role models are more career salient than those who do not have female role models. (Rossi, 1965; Douvan, 1976; Lentz, 1980; Tidball, 1980) Female students in single-sex colleges are more career salient than female students in coeducational colleges. (Lentz, 1980; Tidball, 1980; Solnick, 1995) Female college students who go into non-traditional, professional careers have female role models. (Solnick, 1995)

     Researchers have presented further evidence related to these questions, which suggests the following: Female role modeling is a significant factor in female students' achievement motivation. (Rossi, 1965; Douvan, 1976; Lentz, 1980; Tidball, 1980; Smith & Erb, 1986; Anderson & Ramsey, 1990; Solnick, 1995) There is a positive relationship between female achievement and affirmative action. (Tidball, 1973, 1980; Berger, 1990; Canes & Rosen, 1995) One of the barriers to female achievement is lack of professional, female role models. (Rossi, 1965; Betz & Fitgerald, 1987) There is a positive relationship between females' attitudes and female role models. (McLure & Piel, 1978; Basow & Howe, 1980; Walker, 1984) Female role models affect female earnings. (Angle & Wiseman, 1981; Gerhart, 1990; Fuller & Schoenberger, 1991; Ehrenberg, 1995; Rothstein, 1995)

     Furthermore, researchers have provided evidence that suggests the following: There is a positive relationship between internal locus of control and persistence. (Strain, 1993) Internal locus of control and achievement are positively related. (Lewin & Stephens, 1994) Internal locus of control is a significant factor in females' achievement motivation. (Teglasi, 1978) There is a positive relationship between locus of control and motivation. (Stone & Jackson, 1975) Locus of control and modeling are positively related. (Stone & Jackson, 1975) There is a positive relationship between locus of control in females and female competitors. (Teglasi, 1978) Women have stronger internal locus of control when in competition with women than with men. (Teglasi, 1978) Locus of control and cognitive ability are positively related. (Fry, 1975) There is a positive relationship between change in observers and role models. (Schwartz & Hawkins, 1965; Myrick, 1969; Marlatt, Jacobson, Johnson, & Morrice, 1970; Duke & Frankel, 1971; Sarason, Ganzer, & Singer, 1972; Stone & Stebbins, 1975; Stone & Gotlib,
1975; Dowling & Franz, 1975)

     Thus, there is both indirect and direct evidence related to the question of the relationship, in female students, between internal locus of control and female role models in coeducational settings. Strengthening the internal locus of control in female students may be related to improving achievement of female students and may point towards a solution for the problem of female under-achievement and under-representation.

     This present study was undertaken to answer the specific questions: Is there a relationship between having a female role model and internal locus of control in female students? Is there a relationship in female students at coeducational colleges between internal locus of control and identifying a female role model? Is there is such a relationship, could internal locus of control in female students in coeducational colleges be strengthened by increasing opportunities for students to have female role models? Would identifying female role models strengthen the internal locus of control of female students and improve the students' achievement motivation? The focus of this research was on the relationship between internal locus of control in female college students at a coeducational college and their identification of female role models.

DEFINITION OF TERMS

     The following are definitions of terms used in this study:

Achievement Theory

     This study defines achievement theory as that which incorporates "
. . . features of both cognitive (expectancy-value) and the mechanistic traditions in the study of motivation." (Weiner, 1979: 1790). Achievement theory includes "expectancy value", which concerns how hard one works at achievement as determined by one's success and their valuing of that success. Achievement motivation involves determination to accomplish something difficult, to organize, to master objects, people, or ideas. (Weiner, 1979) In this theory of motivation, "individuals classified as
high versus low in achievement needs exhibit opposing risk preferences when given tasks differing in perceived difficulty . . . . persons labeled high in achievement needs are predicted to exhibit different risk-taking behavior than persons low in achievement needs." (Weiner, 1986:10)

Career Salience

     This term is defined in this study as (a) the degree to which a person is career motivated, (b) the degree to which an occupation is important as a source of satisfaction, and (c) the degree of priority ascribed to the occupation among other sources of satisfaction. (Lentz, 1980: 29; Masih, 1966) Career salience is a factor which is predictive of career orientation and probable career success. (Lentz, 1980)

Locus of Control

     This study defines "locus of control" as that which identifies the extent to which a person perceives that events in her life are contingent upon her own behavior or her own characteristics. (Rotter 1966) Locus of control is operationally measured in this study via Rotter's Internal-External Locus of Control Scale. (Rotter, 1966)

     A person with internal locus of control believes she has control over reinforcing events in his or her life. (Stone & Jackson, 1975) A person with internal locus of control attributes change to herself and to her actions. She believes and acts as though she controls her own future and sees herself as an effective agent in determining the occurrence of reinforcing events. (Stone & Jackson, 1975)

     A person with external locus of control attributes change to an external source. She believes, reports, or acts as though forces beyond her control (such as fate, chance, powerful others, social constraints, or instructions) are important factors in determining the occurrence of reinforcing events. (Stone & Jackson, 1975) A person with external locus of control perceives that reinforcement follows some action of her own but is
not entirely contingent upon her action. She believes that reinforcement is "the result of luck, chance, fate, as under the control of powerful others, or is unpredictable because of the great complexity of the forces." (Rotter, 1966: 227)

Role Model

     This term is defined in this study as an adult female who has qualities or professional skills that a female college student " . . . admires and wishes to emulate . . . The process of observing can take place without direct interaction between the observer and the role model. It is possible that the role model may not be aware of her . . . influence on another person." (Anderson and Ramsey, 1990: 183)

     A role model is defined by social learning theory as an example to imitate, whom a young person perceives as similar to herself. "The social learning viewpoint  states that children imitate same-sex models more than opposite-sex ones due to . . . perceived similarity to the model." (Basow & Howe, 1980: 559) Almquist and Angrist define role models in terms of human examples whom young people can imitate and from whom young people can receive reinforcement. Almquist and Angrist contend that role models go beyond providing simple technical 'how-to' information and set norms and values, provide recognition and reward for achievements, and orient behaviors on a certain course. (Almquist & Angrist, 1971)

     Bell defines role modeling in terms of two processes: identification and interaction. Identification with the role model by the individual can involve any or all of the following: perceived similarity between the individual and role model, imitation by the individual of the role model, and assimilation by the individual of the role model's attitudes and values. Interaction can occur in different spheres of life. (Bell, 1970)  

 Social Learning Theory

     In this study, this theory is defined in terms of individuals' goals, expectancies, and social reinforcements. "According to social learning theory, man's behavior is determined by his goals. Behavior is always directional. An individual responds with those behaviors that he has learned will lead to the greatest satistfaction in a given situation." (Rotter, 1971: 58)

     Rotter's social learning theory proposes that the probability of a behavior occurring is related to the an individual's expectancy that the behavior will gain reinforcement and that the reinforcement has value to
the individual. (Strain, 1993)

     Bandura defines social learning theory in terms of personal and environmental interactions. "In the social learning view, people are neither driven by inner forces nor buffeted by environmental stimuli. Rather,
psychological functioning is explained in terms of a continuous reciprocal interaction of personal and environmental determinants." (Bandura, 1977: 11-12) Bandura contends that the issue of the locus at which behavioral determinants operate relates to reinforcement influences and to environmental stimuli. (Bandura, 1977)

SIGNIFICANCE OF THE STUDY

     Locus of control is a valuable construct in studying academic persistence and motivation. Strain states ". . . other than identifying locus of control as an important motivational factor, research has revealed little about the complex relationship between student motivational behavior and persistence in college." (Strain, 1993: 7) Evidence suggests that persons with internal locus of control are more cognitively able, mentally aware, better predisposed to learning, and more motivated than persons with external locus of control. For example, studies suggest that internal and external subjects differ in cognitive ability and mental awareness. (Stone & Jackson, 1975; Erlund, 1984)

     Internal persons appear to possess more information regarding their status in certain institutions. (Seaman,1963; Erlund, 1984) Internal persons show a greater readiness to perceive relevant information. Stone & Jackson, 1975; Erlund, 1984) Internal persons have been found to be more prepared to seek necessary information in experimental tasks than external persons. (Lefcourt & Wine, 1966; Erlund, 1984)

     Rotter addresses the concept of locus of control and contends that persons with internal locus of control show more overt striving for achievement than persons with external locus of control, who feel that they have little control over their rewards and punishments. (Rotter, 1971; Joe, 1971; Erlund, 1984) Deci, Weiner, and Parsons see internal locus of control as a facilitator of achievement in relation to attributions. (Deci, 1975; Weiner, 1972, 1978; Parsons, 1983) Lefcourt contends that students who believe they can influence the outcome of their work are more likely to be motivated in academic studies. (Lefcourt, 1981, 1983; Strain, 1993)

     An extensive body of literature supports the premise that students with internal locus of control will show higher achievement motivation than students with external locus of control (Coleman, 1966; Parsons, 1983; Weiner, 1978) and the premise that individuals who score high on achievement motivation assume personal responsibility for their work and attribute success to something they personnally do, rather than to luck or ease of task. (Crandall, V. J., Katkovsky, & Preston, 1962; Crandall, V. J., Katkovsky , & Crandall, V., 1965; Coleman, 1966; Parsons, 1983; Weiner, 1978; Strain, 1993)

     Joe reports that his survey of studies on locus of control supports

     . . . the hypothesis that internals not only will show more initiative
     and effort in controlling their environments but also control their
     own impulses better than externals . . . . it appears safe to
     conclude that internals, in contrast to externals, would show a
     greater tendency to seek information and adopt behavior patterns
     which facilitate personal control over their environments
     . . . . (Erlund, 1984: 20; Joe, 1971: 627)

    
     Roueche and Mink further address the value of internal locus of control in individuals:

      . . . they have a higher self-concept and are generally better
      adjusted, more independent, more achieving, more realistic in their
      aspirations, more open to new learning, more creative, more flexible,
      more self-reliant, show more initiative and effort in controlling the
      environment, are less anxious, have higher grades, show more interest
      in intellectual and achievement matters, etc. . . . It would seem
      reasonable then to try and aid people in changing to a more internal
      orientation, to help them realize the contingencies between their own
      behavior and relevant aspects of their environment - - thus increasing
      the efficiency and effectiveness of their behaviors. (Erlund, 1984:
      19-20; Roueche & Mink, 1976: 9)

     Further, internal persons seem to be more cognitively active. (Stone & Jackson, 1975; Erlund, 1984) Evidence suggests that external and internal subjects differ in their mental awareness of challenges inherent in certain tasks (Lefcourt, 1966) and of opportunities. (Stone & Jackson, 1975)

     . . . . Since the control dimension is usually measured by scales
     stressing academic interests, it would seem likely that learning
     skills and achievement behaviors would be highly related to control.
     . . . in investigations concerned with learning and achievement-related
     variables, the control construct allows some predictions when the
     materials are relevant to the subjects' goal strivings. However,
     successful predictions in this area were found only in male samples.
     . . . Perceived control, as need achievement, may be less useful for
     predicting females' achievement behaviors than it is for males'.
     More investigations including sex as a variable are necessary.
    (Lefcourt, 1966: 213-214; Erlund, 1984:20)


     McKay asserts that ". . . McGhee and Crandall (1968) have shown,
that the locus of control construct does successfully predict academic performance, with internal students being more successful than externals." (McKay, 1978: 22; McGhee & Crandall, 1968: 232) Furthermore, Teglasi finds evidence that suggests that internal locus of control in female students is stronger when they are in competition with other female college students than with male college students. Teglasi finds evidence of "fear of success" in female college students in competition with male college students. (Teglasi, 1978) (See Review of Literature section, Teglasi)

     In summary, evidence suggests that persons with internal locus of control are more independent, cognitively able, mentally aware, predisposed to learning, and motivated than persons with external locus of control. Evidence also suggests that females may experience "fear of success" in competition with males, but that internal locus of control in females may be positively affected by other females. Evidence further suggests that achievement motivation in younger females is positively effected by female role models.

     Since increasing the internal locus of control for female college students would tend to be associated with better academic performance, information concerning the relationship between internal locus of control in females and the presence of female role models is of value. Therefore, this study examines the relationship between internal locus of control in female college students at a coeducational college and whether these students identify female role models.

                    EMERGENCE OF RESEARCH QUESTIONS


     The research questions that are addressed in this study emerged as the researcher conducted ethnographic, historical studies at the University of Mary Hardin-Baylor from fall 1992 through spring 1995. The first research looked at the years from 1886-1972, during the female-only days of the university. The researcher studied University of Mary Hardin-Baylor students from the female-only days by reading published works by former students and faculty on the female-only history of the college.

     Further, the researcher surveyed the entire Class of 1945,collected oral histories from female leaders from the female-only days of the school, and audiotaped and transcribed narration from two former students who have lived near the institution and studied its history for 60 years. Patterns of characteristics emerged from these projects. One pattern that emerged from this ethnographic research was the strong influence of female role models, especially that of faculty role models. (See documentation of evidence in appendix.)

     Patterns of characteristics from the ethnograpic, historical studies at the University of Mary Hardin-Baylor from fall 1992 through spring 1995 suggested that these female students had an internal locus of control. Students from the college's female-only days reported that events in their lives were contingent on their own behavior, choices, actions, and personal characteristics, suggesting internal locus of control. Taking responsibility for their own lives, students from the female-only days typically moved into careers for which they had deliberately prepared, suggesting career salience. (See documentation of evidence in appendix.)

     Through reading writings, interviewing, and surveying students of female-only days at this university, the author has concluded that women educated at the University of Mary Hardin-Baylor before the 70s, in general, seemed to have had both an internal locus of control and female role models. Two basic questions emerged from this historical research:

     1. Is there a relationship between the locus of control in female college students' and their identification of female role models?

     2. Does locus of control in female college students tend to become increasingly more internal when they have female role models?

ORGANIZATION OF THE RESEARCH PROJECT

     The two initial questions which emerged from the 1992-95 ethnographic, historical projects generated research questions which this study addressed with the methodology described in Chapter III. A research project was conducted in 1994-1996. This was a study of first-year female college students and their identification of female staff or faculty role models at the University of Mary Hardin-Baylor. Measurement used Rotter's Internal-External Locus of Control Scale and the investigator's questionnaire related to the students' identification of role models. The first purpose of this study was to investigate locus of control as a changeable variable in first-year, female, coeducational college students.
The second purpose of the study was to investigate the relationship of internal locus of control between first-year, female college students and the identification of female role models. This study was built on a review of literature in the fields of education, psychology, sociology, industry, and labor.

REVIEW OF LITERATURE

     Literature that relates to the issues of locus of control in females and the effects of female modeling (of females in general on other females and the effects of female role models on female college students) can be found in the fields of education, sociology, psychology, and labor. Studies in these areas indicate that internal locus of control, self-esteem, career salience, and achievement in females are positively related to the presence of female role models.

Educational Literature

     In educational literature, there are two leading theories and many research projects directly pertinent to this study. Weiner provides a theory of attribution, and Bandura provides a theory of observational learning from models. Studies of many educational researchers support the ideas that having female role models can influence professional choices, career salience, academic and professional success, self-esteem, and attitudes toward women, in both female students and in female professionals. Further, research provides evidence of locus of control as a significant factor in achievement motivation.

     Weiner asserts that people attribute their successes and failures to internal or external reinforcers. An "internal person" attributes successes and failures to her ability or to her effort. An "internal person" attributes her performance to causes for which she assumes personal responsibility. An "external person" attributes her performance to factors for which she has no responsibility and over which she has no control. If she fails, the "external person" assumes that the task was too difficult or that she was unlucky (or both). If the "external person" succeeds, she attributes her success to the easiness of the task or to luck. (Weiner, 1986)

     Bandura's theory of observational learning concerns learning from models. (Bandura, 1969) He asserts that much behavior is acquired through observing and imitating other people. He contends that new patterns of behavior are learned through observing behavior without the observer overtly responding or receiving any reinforcements in the exposure setting. He writes, "Modeling influences . . . can create generative and innovative behavior." (Bandura, 1977: 40-41) He argues that observers watch models performing responses which embody a certain principle. Later the observers behave in a way stylistically similar to the model's behavior, even though the observer is not mimicking the model's specific responses, because the observer has applied what she has learned from the model to a new, but related, situation. (Bandura, 1977)

     Bandura and Walters assert that teachers as role models may have three types of effects on students. The first is the "modeling effect," which involves the student's direct imitation of the model's behavior. The second is the "disinhibitory effect," which involves the student's observing the consequences of the model's actions and consequently choosing behavior in opposition, if the model's observed consequences were undesirable. For example, when female faculty members are regarded with low esteem by school administrators and are not treated as equals, the effect may be to inhibit female students' aspirations toward the teaching profession. The third modeling effect is the "eliciting effect," which involves the increased susceptibility in a student to the influence of the role model. For example, a female teacher who holds high expectations for female students' achievement may have an increased probability of influencing the female students' performance through cues which elicit a positive response in the students. (Bandura and Walters, 1969)

     Studies provide evidence that same-sex role modeling may be especially important for female students. Finn finds that, in England's all-girls schools, girls are superior readers compared with male peers but that reading scores for females in English coeducational schools are about equal to those of male students. (Finn, 1980) Johnson finds evidence of a positive relationship between same sex-role modeling and academic achievement in his studies of students and teachers in the United States, Canada, England, and Nigeria. He finds that females are academically superior in courses taught by females and that males are academically superior in courses taught by males. (Johnson, 1973-74).

     Studies of many educational researchers support the ideas that female role models affect professional choices, career salience, academic and professional success, self-esteem, and attitudes toward women in female students and in female professionals. Douvan asserts that studies of adolescent and college women consistently report that girls and women whose mothers worked outside the home are likely to aspire to higher-skilled and more prestigious jobs than those women whose mothers were full-time housewives. (Baruch, 1972; Tangri, 1972; Douvan & Adelson, 1966).

     Douvan further points out the importance of older women role models in the biographies of successful women. (Douvan, 1976)

     The role of a crucial older women model is prominent in the
     biographies of women successful in politics (Kirkpatrick, 1974),
     poetry (Dickinson, 1984), the academic world (Almquist & Angrist,
     1975; Mead, 1972) and psychoanalysis (Deutsh, 1973; Erickson, 1965).
     In her . . . autobiography Helen Deutsch (1973) writes of her
     experiences . . . , 'I was seeking models with whom I could identify;
     these could only be found among the feminine leaders . . . I found
     two . . . I needed . . . . In this gathering that swarmed with famous
     men, both of these women were treated with significant respect,
     and their speeches had a strong decisive influence on the proceedings.
     I heard these speeches with awe. (Douvan, 1976: 6)

     Douvan reflects on evidence related to the importance of older women role models for younger women at women's colleges. She expresses gratitude, like her undergraduate peers, for the availability of female faculty role models at Vassar. She suggests two important reasons why eastern women's colleges produce a disproportionate number of women who achieve prominent positions in the arts, business, politics and other professional fields. Douvan contends that a serious and intense intellectual climate exists at women's colleges, a climate that has been described by a number of observers. (Newcomb, 1943; Sanford, 1962) That climate contains another reason for the disproportionate number of achieving women at women's colleges. (Adelson, 1962; Katz, 1962)

     . . . modeling is also important in the process that allows these
     women to integrate achievement strivings and the industry and
     discipline needed to realize their talents. Familiarity with
     adult women who have integrated achievement into gratifying
     lives . . . ease . . . anxieties that so commonly accompany and
     inhibit success in women (Horner, 1970) . . . . My colleague was
     particularly affected by women on the Vassar faculty who were
     committed intellectuals and scholars at the same time that
     they . . . led rich, full family lives . . . . the model who most
     impressed me was an unmarried social scientist of international
     reputation . . . . a department chairman with a staff that included
     men. (Douvan, 1976: 7-10)

       Female faculty role models at Vassar impressed Douvan and her undergraduate colleagues with their involvement and commitment. Douvan and her student colleagues were disappointed, however, when they went to graduate school and did not find the same number and availability of female faculty role models. "Graduate school was. . . disjunctive . . . with . . . undergraduate life at Vassar . . . We were concerned about the absence of . . . models. (Douvan, 1976: 7-10)

     Douvan presents Walum's findings in a study of twenty-five female Ph.D.s, who were integrating professional and traditional family activities and who had strong identification with a female role model. In some cases it was a grandmother, aunt, or teacher,. Nearly all of the women spoke with passion concerning the importance these models had in motivating them, not only towards professional careers but also in alleviating the guilt often associated with being a working mother. Walum asks her subjects about books they remember from childhood. Her subjects allude to books about famous women and speak of their identification with the heroines in these books. In a further study of undergraduate women, Walum finds a significant relationship between career aspirations and memories of heroines from books read in childhood. (Douvan, 1976; L. R. Walum, personal communication, 1974).

     Tidball reports from her studies that female role models enhance the career salience of female college students and concludes that women's colleges produce twice as many achieving women as do coeducational colleges. In a paper based on her speech at the American Association of the Advancement of Science, Section on Education, Tidball, professor of physiology at George Washington University, gives her perspective on affirmative action as it relates to academic women. (Tidball, 1973)

     Tidball begins by recounting that girls were not able to attend even public elementary schools prior to the end of the eighteenth century. She asserts, however, that affirmative action existed then as women such as Mary Lyon, Catherine Beecher, and Emma Willard led the way to the founding of the first American colleges for women. Consequently, by the mid-nineteenth century, a number of institutions had been established in which women could participate in studies leading to the Bachelor of Arts degree. Most of these institutions were for females only, although a few were coeducational. By 1870, women comprised one-fourth of all students enrolled in institutions of higher learning. Tidball states that affirmative action in institutions of higher learning has emphasized hiring, paying, and promoting professional women to enhance opportunities for women faculty and administrators. Tidball contends that it has become recognized that the educational climate must be improved for women students and for women who are professionally engaged in the educational process. (Tidball, 1973)

     As more is understood of human development, it will become
     increasingly apparent that the two are related, that the nutritive
     environment for women undergraduates is closely related to the
     environment that prevails for women faculty and administrators,
     and that one cannot expect a high return on the investment in
     women students if institutions are uncommitted or hostile to
     women academic professionals. Thus, the potential effects of
     . . . affirmative action plans are much broader than might appear
     superficially, reaching beyond the individuals specifically involved
     to students and to . . . society as well. (Tidball, 1973: 131)  

     Tidball asserts that the period of late adolescence and early young adulthood is pivotal in "actualizing the potential that is genetically inherent and which was influenced early in life." (Tidball, 1973: 31). She contends that in this period, individuals develop a firm identity and "the capacity to become involved with others, to explore one's inner self, one's . . . commitments." (Tidball, 1973: 131) She argues that according to developmental theory, during adolescence and young adulthood, individuals reevaluate, integrate, and redirect past identities (positive and negative).

     But far too often the young woman, at this crossroad in her life,
     feels a conflict between being accomplished and being female. . .
     If she denies her newly found capabilities, the waste has begun.
     If she accedes to them, she is beset by . . . a fear of success.
     . . . Yet the college years, when this pressure is so strong,
     are critical to the total development of a woman's talent and to
     the momentum and self-assurance necessary to subsequent achievement.
     The apparent dichotomy cannot be taken lightly. . . . The young woman
     must be able to view competence in a . . . spectrum of options at a
     time when she is receptive and capable of making midcourse adjustments.
     Men do not face these same problems . . . (Tidball, 1973: 132)

     Tidball states that due to her understanding of developmental tasks of young women, she studies career-successful women and their undergraduate origins in order to find clues to the composition of a supportive environment during a pivotal direction-setting time in the lives of female college students. She selects at random 500 students from three editions of Who's Who in American Women. Tidball determines, from 1,500 "career-successful" female students over a span of five decades, how many graduated from women's colleges and how many graduated from coeducational colleges. (Tidball, 1973)

     When the number of achievers/1,000 women graduates/decade was
     calculated for the two types of colleges . . . . The paired
     difference calculation indicates that the greater achievement
     output from the women's colleges compared with the coeducational
     colleges is highly significant (p <0.005). The overall comparison
     for the five decades indicated approximately a two-fold difference
     in achiever output for the two types of colleges. (Tidball, 1973: 132)

     Tidball questions why these results are found. She addresses the effect of the presence of female role models on female students in the educational environment. "The influence of women faculty and administrators present in the environment, as well as guest lecturers whom the college places in prominent view, cannot be underestimated in the developmental process of young women, especially at this turning point in life." (Tidball, 1973: 133) Tidball compares the number of women faculty to women students in the female-only and coeducational colleges in her study. " . . . there were almost twice as many women faculty/1,000 women students in women's colleges as in coeducational colleges." (Tidball, 1973: 133)

     Tidball concludes that the results of her study suggest that

     . . . the significantly greater number of women faculty/students at
     women's colleges account in part for the statistically greater
     output of achievers at women's colleges . . . . the number of women
     faculty and the number of women achievers were highly and positively
     correlated; their dependence was not a matter of chance . . . .
     (Tidball, 1973: 133)

     Tidball discusses the implications and impact of her findings . . . .

     These findings have considerable impact. While many educators have,
     for some time, thought the presence of role models was beneficial for
     young women, these data clearly indicate that the development of young
     women of talent into career- successful adults is directly proportional
     to the number of role models to whom they have access. Whether they
     are students in coeducational institutions or women's colleges, a
     continuum exists which is predictive for achiever output. Any future
     affirmative action should, therefore, consider role models.
     (Tidball, 1973: 133)

     Having examined the correlation between 1,500 career-successful women and the number of female academic professionals, Tidball attributes this high career salience among graduates of women's colleges to the greater number of female role models at women's colleges. (Tidball, 1973) However, Lentz, questions the results of Tidball's research since the procedure used to select women was whether or not they were listed in Who's Who of American Women (a biased sample according to Lentz) and because a positive orrelation between the number of career achievers and the number of female faculty, in Lentz' view, were interpreted as having a cause-effect relationship. Lentz reviews the literature by looking at three important variables: the impact of female colleges, the impact of female professors as role models, and female socialization. She is not convinced by the results of previous research concerning these three variables. (Lentz, 1980)

     Lentz studies the career salience (the degree of career motivation) of girls who chose female colleges and who chose coeducational colleges to see if those who chose female colleges were more career salient before they enrolled than those who chose coeducational colleges. She pursues her study by administering questionaires given to female students at the beginning of their freshman year.

     Lentz studied 1818 subjects. They were volunteers, who were not randomly selected. There was a 65% rate of response of questionaires by volunteers. Subjects were from five women's colleges and nine coeducational colleges, which were small, private, liberal-arts, non-denominationally affiliated colleges in the Northeast. All the colleges had yearly fees of over $3500 and offered similar programs. In her causal-comparative research, Lentz studied two existing groups, girls who enrolled in female colleges and girls who enrolled in coeducational colleges. (Lentz, 1980)

     Lentz's study involved ex post facto research in that career salience was looked at as a possible cause of choice of college. Lentz theorized a direction, a causality in her research. Lentz found that females who chose female colleges were more career salient before going to college than females who chose coeducational colleges, and that females who chose females colleges had more career-oriented reasons for their choice of college. Lentz concluded that students who chose female-only colleges did so out of the conviction that the female-only colleges more positively reinforced career salient attitudes and goals. Lentz's study did not discount earlier research concerning what affected career salience in female college students. In her study she simply found that another factor, that of career salience before college enrollment, affected the college choices of females. (Lentz, 1980.)

     Tidball's research analyzes the differential productivities of women's and coeducational colleges,in her study of 1,500 women randomly selected from Who's Who of American Women. (Tidball, 1980) She justifies the source for her random sample, ". . . Who's Who in American Women . . . listed some 24,000 women in thirty-four broad fields who were included by virtue of their achievements rather than of their titles. . . . As a research instrument, this wider range of listings minimizes bias in terms of career-field accomplishments . . . " (Tidball, 1980: 506) Tidball refers, to the research of Oates and Williamson, which utilizes ". . . Who's Who in America for the basis for identifying achievers, to assess differences in career-field participation by graduates of women's and coeducational colleges . . . verifies the outstanding record of the women's colleges in graduating women of accomplishment." (Tidball, 1980: 516, Oates & Williamson, 1978)

     As Tidball discusses the design, results, and conclusions of her research, she explains that the ratio of women achievers to women graduates, or women enrolled, is plotted on a regression line against the number of faculty to women graduates or women enrolled. A positive correlation coefficient is found to be positive and is statistically highly significant. On the regression line, data from the women's colleges (which she finds are twice as productive of achievers and have a women faculty/women student ratio twice as large as the coeducational colleges) tended to cluster toward the upper portion of the regression line and coeducational colleges toward the lower portion of the regression line. She asserts that this statistically highly significant dependence between the two variables indicates that this relationship is very unlikely to occur on a chance basis. The substantial importance of women faculty to achiever productivity can be inferred with strong confidence.

     Tidball attests that when a correlation coefficient is calculated for the relationship between women achievers and men faculty, the obtained value indicates no statistically significant correlation. She contends that the number of men faculty neither enhances nor detracts from the output of women achievers. Tidball examines her data relative to the correlation of men students and women achievers. She relates that this correlation is negative, which she contends means that the larger the proportion of men students on campus, the less likely it is that women are cited for career achievement. (Tidball, 1980)

     In her work on the relationship between baccalaureate origins and career patterns, Tidball uses the Doctorate Record File since ". . . this registry is large, complete, and sub-categorized according to six broad fields of career-related accomplishments. From fifty-three years of data, the ratio of the number of women receiving a doctoral degree to the number of women graduates for each field is calculated separately for graduates of women's and coeducational colleges." (Tidball, 1980: 515, 1978) Tidball contends that graduates from women's colleges are more than twice as likely to have received research doctorates in all fields as are women graduates from coeducational colleges. (Tidball, 1980, 1976)

     Tidball concludes from her research that although women students in women's colleges have extensive opportunities for leadership and are encouraged to study in non-traditional fields, these advantages may be derivatives of high expectations held by the large number of adult women in a women's college who themselves exemplify achievement. Further, she asserts that while women faculty are more supportive of issues that concern women than are men faculty, the proportion of women faculty in all types of institutions, except the women's, is so small that their impact on women students is severely constrained. (Tidball, 1980, 1976, 1974) Thus, she contends that

     . . . the importance of the number of women faculty, in relation
     to the faculty population and to the number of women students,
     emerges and reemerges as a major influence in the development of
     women students who subsequently become achievers . . . . the
     overwhelming majority of women students are to be found in the
     coeducational settings, but we know they have not flourished there
     in the past in numbers commensurate with their presence. Studying
     various aspects of the women's colleges and their graduates has
     provided a glimpse of what women can become when certain environmental
     conditions obtain, most particularly a high women faculty/women
     students ratio. (Tidball, 1980: 516-517)

     Solnick reaffirms Tidball's contention that women's colleges promote achievement, particularly in typically male fields. Solnick contends that

     . . . women's colleges have been found to produce two or three times
     as many natural science Ph.D's, medical school entrants and
     "achievers (identified by inclusion in Who's Who of American Women)
     per capita as coed schools, controlling for size, selectibility,
     and per students expenditure (Tidball 1980, 1985, 1986). Successful
     women's college graduates are more likely than their counterparts
     from coed schools to be found in fields that are traditionally male
     . . . . Tidball identified the percentage of female faculty as the
     crucial factor leading to female achievers. (Solnick, 1995: 506-507)

     Walker also reported on the effects of role models on female students. Interested in interaction and identification in female professional and career role-modeling, Walker designed a study in which freshmen girls were asked to participate in a project which examined the effect of female role models on the girls' career choices. This experimental study involved a homogeneous, randomly-sampled treatment group and a homogenous, randomly-sampled control group. It used a pretest-posttest control group design with random assignment. A random sample of incoming freshmen girls participated in this study. Role models were provided for the female freshmen treatment group in two ways: through interaction with female faculty advisors and through exposure to female career role models in a course on career exploration. At the end of the first year, data was collected from a treatment and from a control group. This data indicated significant changes in self-esteem and in attitudes toward women in the treatment group but not in the control group.

     Walker's results of the first year of the study of the treatment and control groups were reported from attitude surveys: Coppersmith Self-Esteem Inventory, Attitudes toward Women Scale, Life Style Index, and Occupational Status Inventory. Data from the first year of the study, using these measures, indicated significant positive changes in self-esteem and in attitudes toward women in the treatment group but not in the control group. However, when some of these same subjects were surveyed again after the freshman year on the same instrument that they used as freshmen, the results were not as significant. Realizing the value of longitudinal data, Walker surveyed subjects from the original study at the end of four years and did not find the same significance. In the original project, fifty-seven were in the treatment group and thirty-four were in the control group. This was not a large number, and there was significant attrition. Only 59% participated in the study at the end of the fourth year. At that time, analyses of variance indicated significant increases in scores on all measures with no significant differences between the two groups, although the most interesting difference between the two groups was that girls from the treatment group were still more likely to name females as role models than those in the control group after the fourth year. (Walker, 1984)

     Walker and Kuk followed-up, in 1987, these same groups of women, studied originally during their freshman year in 1978, in a longitudinal project designed to investigate the career development of women. Of the 113 surveys mailed, 47 women responded and 45 responses were usable. The 1987 study indicated no significant positive changes in self-esteem and in attitudes toward women. (Walker & Kuk, 1990: 22-23) It was only at the end of the first year of the 1978 study that data indicated significant changes in self-esteem and in attitudes toward women in the treatment group.

     Although in Walker's study the effect of role models on female students is significant only in the students' freshmen year, Walker's study is among those research projects that support the idea that a positive relationship exists between female role models and females students' self-esteem and career salience, especially during the first year of college. (Walker, 1984) An important body of educational theories and research exists which supports the idea that a positive relationship exists between students' career choices, academic and professional success, and attitudes toward women and female professionals. (Lentz, 1980)

     Further, academic achievement is related to the construct of locus of control in educational literature. Strain studies locus of control as a predictor variable related to academic persistence and achievement motivation. She asserts that other educational researchers predict that

     . . . students with internal locus of control will persist longer
     in college. Mink & Watts, 1977; Roueche, J. Mink, Fisher, & Linquist,
     1978) . . . the findings of previous studies . . . warrant the
     conclusion that locus of control should be prominent in any current
     examination of motivation. Further, research findings accumulated
     over the last 20 years affirm the importance of locus of control as
     a factor in the motivation of students (Coleman, 1966; Crandall,
     1968; Deci, 1975; Weiner, 1972, 1978. Strain, 1993: 6)

     Strain contends that locus of control is a valuable construct in studying academic persistence and motivation. " . . . other than identifying locus of control as an important motivational factor, research has revealed little about the complex relationship between student motivational behavior and persistence in college." (Strain, 1993: 7) Strain asserts that the process of motivation is complex but clearly related to locus of control. Strain stresses that studying student motivation is important in that it is linked to student retention and achievement. Strain states that lack of motivation is "associated with the withdrawal of academically unprepared students from college (Roueche, J., 1968; Astin, 1975; Tinto, 1985).

      She further asserts that although factors of intrinsic motivation are a basis for students' persistence behavior, only one concrete, measurable motivational factor, that of locus of control, has been identified in research. Strain asserts that the construct of locus of control has been particularly valuable in studying academic persistence.

     . . . . research conducted in the late 70s . . . conducted at
     multiple institutions with large numbers of students, reported
     findings that locus of control was related to persistence
     (Roueche, J. & Mink, 1976a; Mink & Watts, 1977; Roueche, J., Mink,
     Linquist, & Fisher, 1978). All . . . studies found that students
     who were more internal persisted at higher rates. (Strain, 1993: 31)

     The studies Strain discusses contribute a major finding related to locus of control as a factor in persistence. This finding defines an aspect of weak motivation and identifies " . . . a motivational factor which was amenable to intervention (Mink & Watts, 1977; Roueche, J. & Mink, 1976b; Mink & Watts, 1977). (Strain, 1993: 31) Strain contends that motivational theory provides a basis for associating motivational factors with persistence behavior.

     Motivation describes forces that energize a person and direct
     activity; therefore, persistence is believed to be a behavioral
     pattern that forms a basis for inferences about motivation.
    (Maehr & Archer, 1985) Motivational theory aims to account for
     changes from one activity to another. Intrinsic motivation relates
     to the character of the motivation that is described by goal
     commitment. Thus, an examination of the principles of motivational
     theory could assist in identifying factors of motivation that may
     be basic to student persistence. (Atkinson, 1964; Atkinson &
     Feather, 1966) (Strain, 1993: 30-31)

     Strain studied the effect of locus of control among other possible predictors of persistence in low-achieving students in a two-year college. She examined a sample of 313 students in a developmental studies program enrolled on the basis of cutoff scores from institutional entrance examinations. Locus of control was measured by the adult Nowicki-Strickland Locus of Control Scale. Achievement motivation was measured by a work and family questionnaire, composed of three factors: work, mastery, and competitiveness. Among the possible predictors of persistence were locus of control, achievement motivation, age, sex, race, family income, educational objective, major, intended duration of schooling, and college grade-point average. From the study emerged three significant predictors of persistence: first semester college grade point average, locus of control, and race. A series of t-tests revealed statistically different means on the locus of control scale between female and male subjects. Females were found to be more internal than males.

     Educational research provides evidence of locus of control as a significant factor in achievement motivation. Evidence of locus of control as predictive of academic persistence and role modeling as effectual of academic success is also found in educational literature. Further, there is evidence that role modeling by female professionals affects self-esteem, attitudes toward women, professional choices, career salience and professional success in female students.

Psychological Literature

 
     Studies in the field of psychology provide evidence of locus of control as a changeable variable which can be affected by modeling. Studies in this field also address the relationship of modeling to change, the relationship
of models to locus of control, the relationship of investigation of knowledge (cognitive inquiry) to locus of control, and the relationship of locus of control in women to their competitors. Studies in counseling psychology investigate the use of models to motivate change by observational learning. (Myrick, 1969; Sarason, Ganzer, & Singer, 1972; Stone & Stebbins, 1975)

     Modeling is found positively related to motivation to change in counselees in many counseling categories. In counseling sessions, modeling procedures for counselees are found positively related to motivation in counselees to increase talk time on specific topics (Duke & Frankel, 1971); to increase referring to themselves (Myrick, 1969; Stone & Stebbins, 1975); to increase admitting problems (Marlatt, Jacobson, Johnson, & Morrice, 1970); to increase statements about their feelings (Schwartz & Hawkins, 1965); and to increase self-disclosure (Stone & Gotlib, 1975).

     Dowling and Frantz report the results of a study designed to test
the hypothesis that a facilitative model, one that communicates empathy, respect, and genuineness, enhances imitative learning by an observer.
The subjects, seventy-two college students, were randomly assigned to eight experimental and two control groups, in which the level of facilitative conditions and attitudinal posture of the group leader model were varied. Facilitative models generated significantly more imitative learning than control groups. The results provide evidence that ". . . if the counselor offers certain faciliative conditions . . . with a client, the client will be come more self-directing, more open, and flexible . . . . " (Dowling & Frantz, 1975: 263)

     Numerous studies in the field of psychology have investigated internal locus as a changeable variable and as a variable which can be affected by modeling. Stone & Jackson explored the relationship between locus of control, modeling, and instructional effectiveness in a study using university students, divided into "internals" and "externals" according to scores on Mirel's factor of "personality internality." (Mirels, 1970)       Subjects were randomly assigned to experimental and control groups: instructional (subjects instructed to be highly concrete in talking about experiences); modeling (subjects exposed to a model who discussed experiences in highly concrete fashion); and control (subjects given general
instructions in talking about experiences). All subjects participated in a monologue, after treatment, which was scored for concreteness. Results indicate that modeling is effective, and that internal-modeling subjects
are more effective in concretely talking about experiences than internal-instructions subjects. (Stone & Jackson, 1975)

     Fry relates the results of a study designed with subjects engaged in
an occupational information and vocational exploration study. Subjects were given varying degrees of control over the programming of their exploration, when they were assigned to groups in which high, low, and intermediate levels of external control were manipulated. High cognitive inquiry, high "internal" subjects made significant gains in all treatments but learned most under a high degree of their own control of information studied. High cognitive inquiry, low "internal" subjects gained significantly more under
a high degree of external control of information studied.

     In Fry's study, among the high "internal," high cognitive inquiry subject types, those in the treatment group, in which subjects had a high degree of control themselves over information studied, gained significantly
more than any other subjects in treatment or control groups. These subjects were far superior to other groups and demonstrated the most gain of all subject-type groups in the experiment. Consequently, Fry observed that the
high internal high inquiry subject types had a better predisposition to learning, and, therefore, their gains were superior in all treatment groups. (Fry, 1975)

     Teglasi studied locus of control in 144 female undergraduates,
randomly selected from psychology, sociology, and education classes
at a community college. Pretests and posttests were used in this experimental research. Similar test booklets were passed out to all subjects who are only told they were participating in a study investigating attitudes and creativity. Subjects were randomly assigned to treatment and control groups. Subjects were asked to indicate causal attributions for success or failure outcomes. Working in pairs, one half of the subjects cooperated with either a male or a female partner, while the other half competed with a male or female opponent. All female subjects in the study were pretested on achievement motivation and sex-role orientation. (Teglasi, 1978)

     Teglasi found that females who valued the traditional feminine roles were more self-effacing in causal attributions than non-traditional females. She found that females who valued achievement, like male counterparts, were more self-enhancing following failure.  Teglasi found, however, that, following success in competition against male opponents, females who scored
high in achievement motivation were less self-enhancing than those who scored low. This finding is consistent with the view that females who are highly achievement motivated also fear negative consequences associated with
success, especially in competitive settings with men. (Horner, 1968; Tidball, 1973) There has been some evidence that expectancy and attributional biases which derogate women are culturally determined and shared by men and women. (Deaux & Taynor, 1973; Deaux & Emswiller, 1974; Deaux & Farris, 1974).

     Teglasi investigated what determines the causal attributions of female college students. In her study independent variables were the sex-role orientation and achievement motivation of the students. The manipulated
variables were the following: cooperative-competitive condition, success or failure feedback, and male or female partner or opponent. The dependent variables concerned casual attributions. The following hypotheses were confirmed. In terms of intercorrelations of role orientation, women with traditional role orientation have lower achievement motivation than women
with non-traditional role orientation. In terms of role-orientation, non-traditional women are more self-enhancing than traditional women following failure feedback and non-traditional women are more internal for success than traditional women. In terms of achievement motivation, women who score high on achievement motivation are more self-enhancing in the causal attributions for failure but less self-enhancing for success. (Teglasi, 1978)

     From the results of her study, Teglasi concludes that internal locus of control in women is stronger in relation to other women than in relation to men. She concludes that women are internal in their attributions of their success in competition with women, but external in their attribution of success in competition with men. In her study of a female's locus of control in relationship to competitors, Teglasi contends that, in general, women assume responsibility and personal control of their success in competition with women, but do not, in general, assume responsibility and personal control of their success in competition with men. She asserts that women demonstrate stronger internal locus of control in competition with women. (Teglasi, 1978)

     Literature in the field of psychology provides evidence of locus of control as a changeable variable, of the positive relationship of role
models to locus of control, of the positive relationship of cognitive
inquiry to locus of control, of the positive relationship of role models to change, and of the positive relationship of locus of control in women in competition with other women.

 
Sociological Literature

     Many studies in the field of sociology also present evidence concerning the effect of the presence or absence of role models on female college students. Rossi supports the idea of the positive effect of female role models on female college students, after her extensive study of female professional career choices and the barriers female professionals encounter in the fields of science, engineering, and medicine. Sponsored by the Carnegie Corporation, the National Institute of Health, and the National Opinion Research Center, Rossi identifies factors which encourage and discourage female achievement in professional fields, especially in science, engineering, and medicine. In her conclusions, Rossi addresses the issues
of stereotypes and social climate that discourage female professional,
non-traditional achievement by discouraging females to be independent thinkers. She concludes that women in non-traditional professions are more likely to have had female professional role models. (Rossi, 1965)

    Women scientists, engineers, and doctors . . . . womanpower is seen
    as a major reservoir from which the nation may draw to meet the
    continuing critical shortages in these important fields . . . .
    Campaigns to . . . support . . . the college-age woman to enter the
    sciences, engineering, or medicine can only . . . reach and help the
    young woman who is already interested and prepared by a background
    in science and mathematics to take advantage of opportunities offered
    her college. Such women are a tiny minority of their sex, whose
    experiences at much earlier ages have set in motion an . . . interest
    in things generally disapproved of for girls . . .

    Rossi further contends that female college freshmen seldom shift from traditionally female fields to traditionally male fields, such as from

    . . . fine arts to chemistry, or from journalism to engineering,
    except in rare instances. . . . efforts to be . . . effective must
    concentrate on . . . earlier stages of life and must involve . . .
    changes in the rearing of girls. . .(Rossi, 1965; 52-54)

    Rossi insists that stereotypes are persistent, but if we are seriously

    . . . interested in attracting more women into medicine and engineering,
    it is surely worthwhile to correct these . . . and portray the
    contemporary doctor . . . engineer in work contexts typical of
    today rather than yesterday. . . . Young women need exposure to women
    scientists and doctors as models. (Rossi, 1965: 98)

    Rossi addresses the need for a social climate that encourages females to be the kind of independent thinkers that are needed in the fields of science and math. She asserts that

    . . . . results suggest that, on the average, girls develop cognitive
    abilities along somewhat different lines from boys, and that they enter
    adolescence with a style of thinking less appropriate to scientific work
    than that of boys. Although the final interpretation of this sex
    difference awaits further research, what is known is that the key to
    the difference between boys and girls lies in the kind of and degree
    of independence training the child receives in childhood. If a girl
    is encouraged to assume initiative, to solve problems for herself,
    she tends to develop the same analytic abilities as the boy typically
    does. . . . (Rossi, 1965: 115)

    Rossi asserts that behind the fact that few women are represented in science is the difference in cognitive style of male and female, which is affected by the general culture.
  
    . . . which is in turn the result of differences in the way girls are
    brought up compared to boys. If we want more women to enter science,
    not only as teachers of science but as scientists, we must
    encourage the cultivation of the analytic and mathematical abilities
    science requires. To achieve this means encouraging . . . self-reliance
    instead of feminine submission . . . (Rossi, 1965: 117)

    Furthermore, Rossi describes relationships which encourage a girl toward science and other professional fields. ". . . such women, when questioned
in college about the background of their science interests, frequently point
to a particularly important teachers they had, often as early as the third
or fourth grade, who stimulated and challenged them to stretch their minds." She contends that women in non-traditional professions are more likely to have had female professional role models. (Rossi, 1965: 122).

     Fox computed, for several universities, the correlation between the number of female faculty and the number of female undergraduates in major academic divisions: social sciences, natural sciences, humanities, education, and applied sciences. Fox reported a positive correlation from her computation between the number of female faculty and the number of female undergraduates in these major academic departments. She interpreted this result as support for the idea that same-gender role models affect undergraduate choice of major. (Fox, 1974)

     Basow and Howe study role-model influence in college students and give an overview of related literature on the subject.

     . . . a number of studies have demonstrated the importance and
     influence of female role models on the career development of women
     at various ages and educational levels. Mothers have been found to
     be particularly important in influencing their daughters' career
     orientations and attitudes (Almquist & Angrist, 1971; Baruch, 1972,
     Tangri, 1972). Exposure to other female role models through work
     experiences has been found to be influential to the career-oriented
     college women studied by Astin, . . . as well as Almquist and Angrist
     (1971) and O'Donnell and Anderson (1978). Female teachers and college
     influences on female students in regard to achievement and success,
     (Tidball, 1973), goals and values (Douvan, 1976), and subsequent
     professional achievement and productivity (Epstein, 1970; Goldstein,
     1979).(Basow & Howe, 1980: 559-560)

     Based on their review of literature, Basow and Howe predict that women are particularly influenced by female role models in making important life
decisions. Besides the variable of the same-sex subject and same-sex model
in their research on role modeling, Basow and Howe consider another variable in the role-modeling process, that of an individual attitudes regarding sex-role behaviors and stereotypes. Basow and Howe give an overview of the literature on this variable. They find that one basis for feelings of interpersonal cohesiveness is attitude similarity (Frieze, Parson,Johnson, Ruble, & Zellman, 1978).

     For example, persons with traditional sex-role attitudes might
     therefore expected to be more influenced by traditional role models.
     Furthermore, cognitive development theorists emphasize the role
     of sex-role attitudes and stereotypes in the process of information
     categorization and development of sex-role behaviors (Kohlberg, 1966).
     (Basow & Howe, 1980: 560)

     Basow and Howe measure sex-role attitudes with the Attitudes Toward Women Scale (AWS) designed by Spence and Helmreich (1972a). The Attitudes Toward Women Scale involves the degree to which a subject accepts or rejects
sex-role stereotypes and identifies subjects as traditional, moderate, or liberal based on the subjects' attitudes toward rights and freedom for women. Subjects classified as "traditionals" would look more toward standard authority figures as role models. "Moderate" or "liberal" subjects would look more toward nonstandard authority figures. Basow and Howe examine the effect of role models on certain decisions made by undergraduates on the basis of their sex and sex-role attitudes. (Basow & Howe, 1980)

     In Basow and Howe's first experiment they investigated the effect of role models on college choice, high school career choice, and college career choice for college undergraduates. Basow and Howe's second experiment investigated the effect of role models on college career choice for college seniors. Their research project involved sixty-two undergraduates, evenly
divided by sex. Basow and Howe rated the degree of influence of six or eight people on the students' choice of college and career. They also assessed the attitudes of the subjects on sex-roles. In the first of the two experiments, the sexes showed different patterns of influence as did the sex-role groups. In the second experiment, in which only college seniors were used and
matched by sexes on an attitude scale, the researchers found again that females were more influenced by female models in their choice of career than were males. Basow and Howe conclude from their study of college students
that "female models are particularly important for female college students in their career decisions, especially mothers and female teachers." (Basow & Howe, 1980: 571)

     Many studies in the field of sociology provide evidence concerning the effect of the presence or absence of role models on female college students as related to science careers. Smith and Erb report on an experiment in which one group of students were in contact with female scientists and a control group was not. In a pretest survey, the groups were similar in their responses to questions about women in science. In a posttest survey, the experimental group had a more positive attitude toward women in science. Smith and Erb conclude that "the use of women science career models may positively affect both enrollment in science courses by girls entering high school and their personal consideration of a science career." (Smith & Erb, 1986: 673) However, Canes and Rosen challenge Smith and Erb's conclusions. They contend that just because the experimental group in the Smith and Erb study had a more favorable attitude toward women in science, following exposure to women scientists, that did not necessarily mean that the girls in the experimental group were more likely to enroll in science classes. (Canes & Rosen, 1995)

     McLure and Piel report results of a survey concerning females' perception of career barriers for females in science careers. After interviewing college-bound female high school seniors, McLure and Piel, conclude that "one of the major barriers" to women's success in science is that females "perceive that preparation for science careers is too difficult because they lack awareness of successful women scientists." (Canes & Rosen, 1995: 487, McLure & Piel, 1978) From the McLure-Piel survey, Betz and Fitzgerald argue that ". . . there is evidence that the relative lack of female faculty is a deterrent to women's educational and career pursuits, particularly in science (McLure & Piel, 1978) and other pioneer fields (O'Donnell & Anderson, 1978)." (Betz & Fitzgerald, 1987: 70-71)

     Betz and Fitzgerald study the career psychology of women and contend that a significant
    
     . . . barrier to women in education has been the lack of female role
     models and mentors, most notably female faculty within higher
     education (Douvan, 1976). . . The facilitative effects of same-sex
     models and mentors were suggested by Goldstein's (1979) report that
     about 80% of the articles published by recent doctoral degree
     recipients were published by individuals whose faculty advisers were
     the same sex. Tidball (1980) also reported that as the proportion of
     women faculty relative to . . .women students increases, so does the
     proportion of women high achievers in professional life. (Betz &
     Fitzgerald, 1987: 70-71)

     Lafortune asserts that "only an increase in the number of women in scientific careers, and/or the teaching of math and science at advanced levels, will change the masculine . . . image associated with these fields, and encourage more girls to enter them." (Lafortune, 1990: 273) Furthermore, Tilghman argues that "all but the most determined women will tend to gravitate to the environment which is most positive and rewarding, and that tends to be where other women have led the way." (Tilghman, 1993: A17)

     Canes and Rosen assert that young women need female role models because a person is most likely to emulate someone who appears similar to herself in external characteristics. They argue for the need for female role models in light of the traditional role of women in American society. According to Canes' and Rosen's position, the character attributes needed for professional achievement are incompatible with traditional feminine qualities, so that young women with serious career ambitions need female role models to demonstrate that success is possible. Canes and Rosen state that sociologists present strong evidence and arguments concerning the value of female role models for female students, especially in the field of science. However, they urge that further research be conducted, especially as to the effect of female role models on the specific endeavors of female students. (Canes & Rosen, 1995)

     In summary, the effects of same-gender role models have been studied
     in the psychological and sociological literature . . . .the general
     view is that role models affect women's educational and career choices.
     Most of the evidence, however, is based either on case studies or on
     surveys. We have found no attempts explicitly to relate changes in
     the number of female role models to changes in young women's . . .
     careers. (Canes & Rosen, 1995: 488)

Industrial and Labor Literature

 
     The fields of industry and labor also present evidence concerning the positive relationship of female role models to female college students' educational and professional choices and attainments. In April 1995,
The Industrial and Labor Relations Review published a series of papers entitled "Symposium: Role Models in Education." Three papers deal with the importance of female faculty as role models in higher education: "Following in Her Footsteps? Women's Choices of College Majors and Faculty Gender Composition" by Canes and Rosen, "Changes in Women's Majors from Entrance
to Graduation at Women's and Coeducational Colleges" by Solnick, and
"Do Female Faculty Influence Female Students' Educational and Labor Market Attainments?" by Rothstein.

     Canes and Rosen address whether the presence of female faculty in traditionally male science and engineering fields induces female students to enter those fields. They point out this question cannot be answered by looking across departments at an institution, at a point in time, to determine the correlation between the percentage of female majors in a department and its percentage of female students. They suggest that this
question can only be answered by looking at whether changes over time in the proportion of a department's faculty that are women are positively related to changes in the proportion of the department's majors that are women. (Canes & Rosen, 1995)

     Ehrenberg discusses the Canes and Rosen study and describes how they conducted tests using

     . . . several years' data on the fractions of female faculty and
     majors in different fields at a private research university
    (Princeton), a public research university (Michigan), and a
     coeducational liberal arts college (Whittier). Using panel data
     methods, they find no evidence that an increase in the
     fraction of female faculty in a department was associated with
     an increase in the fraction of its majors that were female . . . .
     Why did these authors fail to find any evidence to support the
     "role model" hypothesis? One possibility, they suggest is that the
     importance of role models and mentors may be greater at earlier stages
     in a student's educational career than at later stages. (Ehrenberg,
     1995: 483; Canes & Rosen, 1995)

     Solnick hypothesizes in her study of 2,500 female college students that women who begin in traditionally female majors are more likely to shift to other types of majors if they attend a traditionally single-sex women's college than if they attend a coeducational institution. She hypothesizes, secondly, that women who start in majors traditionally dominated by males are more likely to persist in those majors if they attend single-sex colleges. Solnick tests these hypotheses using data on subjects from eight women's colleges and seven coed colleges. She finds evidence to support her first hypothesis but not her second hypothesis.

     Solnick found evidence to support her theory that women who begin in traditionally female majors are more likely to shift to other majors, if they attend a single-sex women's college than if they attend a coeducational institution. She found that women at single-sex colleges were more likely to leave female-dominated majors than women in coed colleges. Solnick concluded that single-sex schools may benefit female students by providing them more flexibility to move into majors in traditionally male-dominated fields that tend to be more rewarded in the labor market. (Ehrenberg, 1995; Solnick, 1995)

     My analysis of a sample of eight women's colleges and seven  
     coeducational institutions, like other similar studies (Polachek,
     1978; Angle and Wissman, 1981; Lyson 1981; Jacobs, 1985, 1986,
     1989; Fuller and Schoenberger, 1991) shows marked differences
     between the sexes in initial and final choice of major . . .
     women's choices of college major appear to contribute to the
     persistent gender gap in wages. In research by Daymont and Andrisani
     (1984), differences in college majors accounted for between 28% and
     43% of the difference in earnings between men and women. A study
     that distinguished starting and current salaries reported that
     differences in college majors accounted for 43% of the gap in the
     former and 18% of the gap in the latter (Gerhart,1990). (Solnick,
     1995: 512)

     Solnick's study yields support for the proposition that gender compositions of students at a college, coeducational or single-sex, influences the movement of students among majors. However, Solnick found women at single-sex colleges substantially more likely to leave female-dominated majors than women at coeducational colleges. In her study, the women's colleges yielded fewer graduates in female-dominated fields and more graduates in male-dominated fields than did the coeducational colleges. "Since wages for college graduates depend in part on field of study, exploring the determinants of college major may further understanding of labor market discrimination." (Solnick, 1995: 513)

     Rothstein addresses the concern over the low percentage of female university faculty in the United States and, consequently, the possible long term effect on female students' advanced educational and market attainments.

     There is . . . concern that the percentage of female faculty at U.S.
     colleges and universities is too low . . . . a recent article in the
     New York Times (1993) emphasized this issue, and noted that whereas
     about 20% of faculty . . . are women, the percentage of undergraduate
     women is more than double that figure. The question arises as to how
     female students might benefit from an increase in the percentage of
     female faculty. One view is that female faculty act as mentors and
     role models for female students, and thus promote their subsequent
     educational and labor market attainments. (Rothstein, 1995: 515,
     De Palma, 1993)   

     Rothstein studies the question of the influence of female faculty on female, coeducational college students' post-graduate achievements using data from the National Longitudinal Study of the High School Class of 1972. The results of her study show a statistically significant positive relationship between the percentage of female faculty and the probability that female students would attain advanced degrees. Rothstein does not find that the percentage of female faculty has a statistically significant direct effect on female students' labor market earnings. However, the advanced degrees of the subjects does have a positive impact on earnings. (Rothstein, 1995)

     The influence of the percentage of female faculty is especially
     interesting, because it may be that by acting as role models, or
     (indirectly) creating a favorable environment at the institution,
     female faculty may influence the career paths of female
     students. (Rothstein, 1995: 516)

     Ehrenberg, Goldhaber, & Brewer address and extend the question as to whether role models are most important early in an individuals' secondary educational career.

     Canes & Rosen and Solnick's suggestion that role models . . . may
     be more important early in an individual's career than later in
     the career is a theme explored in the paper by Daniel Goldhaber,
     Dominic Brewer, and myself, which addresses the importance of teachers'
     race, gender, and ethnicity in secondary schools. Using data from
     the National Educational Longitudinal Study of 1988, which permits
     researchers to match students with individual teachers, we examine
     how a teacher's race, gender, and ethnicity influence students from
     both the same and different race, gender, and ethnic groups.
     (Ehrenberg, 1995: 485)

     Ehrenberg, Goldhaber & Brewer find teachers race, gender, and ethnicity are more likely to influence teacher's subjective evaluations of their students than to influence what the students learned.

     . . . although white female mathematics and science students' test
     scores did not increase more if their mathematics and science teachers
     were white women than if they were white men, white female teachers
     in these fields tended to evaluate these students more highly than
     did white male teachers. To the extent that these higher subjective
     evaluations are communicated to the students or lead to placement
     of the students in higher "track" classes, these female teachers may
     well be serving as mentors for their female students. (Ehrenberg,
     1995: 485)

     Literature in the four fields of education, sociology, psychology,
and labor points toward the positive relationship of locus of control in females and female modeling (of females in general on other females, and the effects of female role models on female college students). Studies in these areas indicate that internal locus of control, self-esteem, career salience, and achievement in females is positively related to the presence of female role models.

Research Questions

     Two basic questions emerged from the 1992-1995 University of Mary Hardin-Baylor historical study:

     1. Is locus of control in female college students a

changeable variable?

     2. Does female college students' locus of control tend to become increasingly more internal when they are influenced by female role models?

These questions guided the development of seven research questions for the 1994-1996 research project:

     1. Will first-year female students who identify for

themselves a female faculty or staff role model (RM) measure as having a stronger internal locus of control than first-year female students at the University of Mary Hardin-Baylor who do not claim to have a female faculty or staff role model (NM)?

     2. Would a testing of subjects in October, the fifth week of the first semester of a school year, show a statistical difference in locus of control between the RM group and the NM group?

     3. If so, what would be the difference in their locus of control?

     4. Would same subjects in the RM group and in the NM group tested in May, at the end of school year, show a difference in, or a change of direction of, locus of control?

     5. If so, what would be the difference in, or change of direction of, locus of control of the RM and NM groups from the beginning of the school year to the end of the school year?

     6. Would a second group tested in March, the fifth week of the second semester of the school year, and in May, at the end of the school year, show similar differences in, or change of direction of, locus of control, as those of the first group tested in October and May of the same school year?

     7. What would the relationship be between identification of role models and the direction of change of locus of control over the course of one school year?

Research Design

     To answer these questions, the researcher developed a study to investigate the relationship of locus of control in first-year, coeducational female students who had identified a staff or faculty role model, or not, at the University of Mary Hardin-Baylor. The population of first-year female students was selected because evidence suggested that the influence of role models may be greater at earlier stages in a college student's educational career than at later stages. (Canes & Rosen, 1995)

     In this study of first-year female, coeducational students at
the University of Mary Hardin-Baylor, in mid-October, during the fall semesters of 1994-95, two instruments were used in pretests: Rotter's Internal-External Locus of Control Scale (Rotter 1966) and a questionnaire developed by the researcher. This questionnaire was used by the subjects to identify whether or not they had a role model at the university and to describe their perceptions of the role model's attributes. The term role was defined as an individual who had "skills or qualities" that the student admired and wished "to emulate." (Anderson & Ramsey, 1990:183)

     These two instruments were used again in spring semester pretests in 1995-96. Rotter's Internal-External Locus of Control Scale was chosen because of the reliability and validity of the scale. The researcher's role model questionnaire included a definition of a role model, a section in which the subject marked whether or not she had a role model, and questions about the role model's attributes. The answers to the questions were designed to validate whether or not the subject had identified a role model. A brief section was included for the subjects to add attributes which they perceived important but had been unlisted on the questionnaire. This second pretest, administered at the beginning of the the 1995-96 spring semesters, was to provide reliability.

     In the study conducted in 1994-96, correlated t-tests were used to study the relationship of the identification or not of role models to locus of control in first-year female college students at the University of Mary Hardin-Baylor. The independent variable was the identification, or not, by the students of role models. The dependent variable was the locus of control of the first-year female students as measured by Rotter's Internal-External Locus of Control Scale.

Hypotheses

     1. First-year female college students will develop over a semester, or a year, a stronger internal locus of control, as measured by Rotter's Internal-External Locusof Control Scale, if they have a female faculty or staff role model (RM), than the no role model (NM) group.

     2. Scores of first-year female college students obtained from Rotter's Internal-External Locus of Control Scale, given at the beginning of the first and second semesters, and at the end of the school year, will render
a significant value at the alpha level of .05.

Subjects

     Four mandatory courses with the largest number of first-year female students were selected to obtain the largest sample. All first-year female students in these courses (N=189) were studied. The study of first-year female, coeducational college students at the University of Mary Hardin-Baylor began in the fall of 1994 and was continued through the spring of 1996.

     In mid-October, during the fall semesters of 1994-96, first-year female students (N=95) enrolled in 1994 study skills classes and 1995 public speaking classes were pretested at the university at the end of class periods at the same time of day. A continuation of this study was conducted in the spring semesters of 1995-96. First-year female students (N=93) enrolled in composition classes and mathematics classes at the University of Mary Hardin-Baylor were pretested. They were tested during the fifth week of spring semester 1995 at the end of morning classes. During the last week of the spring semesters 1995-96, all students who were pretested in mid-October during the fall semesters of 1994-95 and in the beginning of the spring semesters of 1995-96, and who were still enrolled at the university, were invited and encouraged to be posttested during the same range of hours in the office of the researcher. The same instruments as had been used earlier in each pretesting were used in the posstesting. The sample population was reasonably homogeneous, with the basic difference of whether or not role models had been identified. There was no controlled "treatment." The comparison groups were those subjects who had identified female staff or faculty role models.

Procedures

     Four mandatory courses with the largest number of first-year female students were selected to obtain the largest sample size. All first-year female students in these courses were studied (N=189). Faculty members teaching these courses were contacted for permission to survey their students. A cover letter was given to the students describing the study. Participants were reassured that their participation was voluntary, confidential, and would not jeopardize them.  The completed surveys and results were kept in the researcher's file cabinet in her locked office.
The students were reassured that the results and conclusions of the study, of which they were to be informed, could be beneficial to them academically
and professionally. This study posed no risks to the participants. It provided only potential benefit to the participants and to society (see Significance of Study section).

     The pretests in this study were conducted in classrooms and the posttests were conducted in the researcher's office at the University of Mary Hardin-Baylor. All students participating in this study in all groups from fall 1994 to spring 1996 were pretested and posttested with Rotter's Internal-External Locus of Control Scale and with the researcher's questionnaire. In this study, correlated t-tests were used to study the relationship of identification of female staff or faculty to locus of control in first-year female, coeducational college students.

     Permission was granted by the university's academic vice president and by participating faculty and students.

DATA ANALYSIS

 
     For purposes of data analysis, the subjects of this study were divided into two groups. Each group completed the measurement instruments during three different testing periods. Therefore, there is a total of six scores in the overall design: two pretests and one posttest per group. The administration periods were in October, March, and May, during the 1994-96
school years.

     Furthermore, as a control measure, a portion of the population completed both the pretest and the posttest, thereby permitting a matched pairs t-test analysis.

     The examination of this data showed the subjects were essentially equal in locus of control in the baseline measurement period (mean 14.29 NM and
14.32 RM).

     The scores from the posttest period in May demonstrated the predicted effects of the identification of a role model on internal locus of control (mean 11.78 NM and 15.76 RM). This data clearly represented that locus of control is subject to change. Thus, this first hypothesis was upheld
(p < .0001, absolute value difference scores-posttest, p < .001 matched pairs posttest, and p < .05-matched pairs difference scores) .

CONCLUSIONS

 
     In each of the 1994-96 pretests and posttests, there is difference
and direction in the results. There is significance at the .001 level in
the posttests.Therefore, evidence provided by this study suggests that first-year, female, coeducational college students who identify a female faculty or staff role model may have stronger internal locus of control
than first-year, female, coeducational, college students who do not identify a female faculty or staff role model. This study also presents evidence
that suggests that locus of control may be a changeable variable and may increase over the first year of college in female, coeducational college students who identify a female faculty or staff role model.

     Further, this study presents evidence that suggests that internal
locus of control may decrease over the first year of college for female, coeducational college students who do not identify a female faculty or
staff role model. The results of this study indicate that for first-year, female, coeducational college students  there may be a positive effect for students who identify a female faculty or staff role model and a negative
effect for those who do not.

     The first hypothesis of this study was supported that first-year
female college students will develop over a semester, or a year, a stronger internal locus of control, as measured by Rotter's Internal-External Locus
of Control Scale, if they have a female faculty or staff role model, than
if they have no role model. The second hypothesis was supported, by the 1994-96 testings, that scores of first-year, female college students, measured by Rotter's Internal-External Locus of Control Scale, given at
the beginning of the first and second semesters and at the end of the school year, will demonstrate a a significant change in locus of control
(alpha .05).

     These scores were compared by correlational t-tests and showed significance (p < .001 posttest, p < .05 matched pairs difference scores, and p < .001 matched pairs posttest). Therefore, this study suggests a factor which may strengthen internal locus of control in first-year female college students, especially in coeducational settings.

CHAPTER FIVE