l [21' I. ii. "if; 2? Mfiestzeda ééfieee Umfivmsiay ‘——— lllLlHHlllllQllilllUljlfllzlllllllllllllllll mat-:axad This is to certify that the dissertation entitled Development Of A Perceived Environmental Control Measure presented by Randall J. Champeau has been accepted towards fulfillment I of the requirements for Major professor _.;25”_, Date 0-12771 MCI! it nu Affimnvhn ‘ MSU LlBRARlES RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. l" *va r-r . is 15? JUN 041999 » P“: 3:28 DEVELOPMENT OF A PERCEIVED ENVIRONMENTAL CONTROL MEASURE By Randall Joseph Champeau A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Michigan State University 1982 [‘\ F~ ? 6/! ABSTRACT DEVELOPMENT OF A PERCEIVED ENVIRONMENTAL CONTROL MEASURE By Randall Joseph Champeau Locus of Control (L of C) constitutes a personality dimension that may be used in conjunction with other variables to explain and/or pre- dict human social behavior. The purpose of this study was to construct a L of C instrument which could be used to measure the expected re— inforcement perceived by an individual if specific types of environmental action are taken in a given situation. The instrument as designed was entitled the Perceived Environmental Control Measure (PECM). The final instrument was comprised of 45 PECM items in three L of C belief orientations; 1) Internal (I); 2) Powerful Others (P); 3) Chance (C). These three belief systems were in turn applied equally across five categories of environmental action; 1) Legal Action; 2) Per— suasive Action; 3) Political Action; 4) Ecomanagement; 5) Economic Action. Subjects were presented an environmental issue summary and asked to respond to the PECM statements as they pertained to the given issue. Participants in the study included sample populations of college students, Sierra Club members and K—S teachers. Results of this study support the proposed relationship between L of C and environmental action taking behavior. The majority of subjects perceived themselves as having some personal control over the stated environmental issues. However, powerful others and chance Randall Joseph Champeau were also identified as agents having some control over the issues. Subjects showed a tendency to feel most in control with the "take it to court" or legal action process. It was further determined that sex, age and/or occupation may have some relationship to perceived control of the issue. College students scored significantly more internal than teachers of the study. Also college females were found to be more internal than college males. The PECM exhibited evidence of reliability, content validity and construct validity. It contains a set of subscales which show po- tential for diagnosing environmental action taking behavior and for evaluating the effectiveness of environmental education curriculum. The time and efforts expended on this study are dedicated with love to two very special people, my parents. They will always be in my mind and on my heart. ACKNOWLEDGEMENTS Most important, sincere appreciation is extended to those friends of mine who provided the support, encouragement and love which made the completion of this study a reality. Dr. Ben Peyton the chairman of my graduate committee deserves a major debt of gratitude. His support and guidance were invaluable. I truly consider him an exceptional advisor, teacher and friend. Appreciation is also extended to Dr. Martin Hetherington, Dr. David Johnson and Dr. Carroll Wamhoff for serving on my graduate committee. Finally I am grateful to Karen Deford and Susan Hazard for the time they spent typing this dissertation. iii TABLE OF CONTENTS Page LIST OF TABLES ................................................ Vi LIST OF FIGURES ........................................ xi CHAPTER 1. INTRODUCTION . . .. ......................................... 1 Locus of Control and Environmental Action ................ 5 Statement of the Problem ...... ........................... 10 Significance of the Study .... ........... . ................ 11 Limitations of the Studyv ................................. 14 Goals and Strategies of the Study ........................ 15 2. REVIEW OF RELATED LITERATURE ................... . ........ . 18 Theoretical Origin of Locus of Control ................... 18 Measuring Locus of Control ... ............................ 22 Environmental Action Categories ...... .................... 32 Relationship Between Environmental Action Taking and Locus of Control ........ . ..... .... ................. 36 Literature Review Summary ... ................ . ............ 43 3. RESEARCH PROCEDURES ...................................... 4S Instrument Design ....................................... 45 Validity and Reliability of the PECM. ..................... 58 Pilot Study I ......... ..... ......... ......... . ........... 61 Pilot Study II ........................................... 61 Field Testing the PECM ............. . .................... . 63 Data Analysis ................................... ..... .... 69 4. RESULTS OFDATAANALYSIS .................... 72‘ Pilot Study I and II Data Analysis Results ............... 72 Field Testing Data Analysis Results ... .............. ..... 76 iv Chapter 5. DISCUSSION OF FINDINGS AND RECOMMENDATIONS .. .......... Discussion of Findings ... ............................. Implications and Recommendations ...................... Summary of Recommendations ............................ Appendix A. Rotter's Internal-External (I-E) Locus of Control Scale .......................... . ..................... B. Pilot Study II Perceived Environmental Control Measure (PECM) ..... ........ .. ........................ C. Field Study (Final) Perceived Environmental Control Measure (PECM) with Both Environmental Issue Summaries . .............. . ..................... . LIST OF REFERENCES ......-. .............. . .............. ... 139 139 159 164 181 LIST OF TABLES Table Page 3.1 Number of Items and Range of Scores for Total I, P, and C Variables in the PECM ..................... 52 3.2 Number of Items and Range of Scores for Total Action Category Variables in the PECM ................. 54 3.3 Number of Items and Range of Scores for Total PECM Scores .......... .... ............... . ............. 55 4.1 Reliability Coefficients for PECM Scoring Variables ... 75 4.2 Pearson Product-Moment Correlation Coefficients for Total I, P, and C Scores with Citizen Action Scores ............ ....... ....... ............... 77 4.3 Pearson Product-Mbment Correlation Coefficients for Total Action Category Scores and Citizen Action Scores ....... .... ............. .... ............. 78 4.4 Pearson Product-Moment Correlation Coefficients for Total PECM Scores and Citizen Action Scores ....... 79 4.5 Reliability Coefficients Of Total PECM Scores for each of the Sample Groups ..... . ..... . ................. 82 4.6 Reliability Coefficients of Total I, P, and C Subscales for each of the Sample Groups .. ............. 33 4.7 Reliability Coefficients of each Environmental Action Category Subscale for each.of the Sample Groups .. .......... ..... .................. . ............ 84 4.8 Matrix of Corrected Item Total and Respective Item Subscale Correlations for the Environmental Issues Class (Group I) ........................ ... ............ 86 4.9 Range of Scores, Means, Medians and Standard Deviations of Total PECM Scores for each Sample Group.. 89 vi Table Page 4.10 Range of Scores, Means, Medians and Standard Deviations of Internal(I) Scores for each Sample Group ........... . ..................................... 90 4.11 Range of Scores, Means, Medians and Standard Deviations of Total Powerful Others (P) Scores for each Sample Group ..................................... 91 4.12 Range of Scores, Means, Medians and Standard Deviations of Total Chance (C) Scores for each Sample Group .......................................... 92 4.13 Range of Scores, Means, Medians and Standard Deviations of Total Persuasive Action Scores for each Sample Group ..................................... 94 4.14 Range of Scores, Means, Medians and Standard Deviations of Total Political Action Scores for each Sample Group ..................................... 95 4.l5 Range of Scores, Means, Medians and Standard Deviations of Total Ecomanagement Action Scores for each Sample Group . ........... .... ..... . ........... 96 4.16 Range of Scores, Means, Medians and Standard Deviations of Total Economic Action Scores for each Sample Group ..................................... 97 4.17 Range of Scores, Means, Medians and Standard Deviation of Total Legal Action Scores for 8 each Sample Group ..................................... 9 4.18 Means and Standard Deviations of Citizen Action 99 Scores for each Sample Group ............... . .......... 4.19 Pearson Product-Moment Correlations for Total PECM Scores with Citizen Action Scores for each of the Sample Groups and for Sample Groups responding to the Solid waste Issue Combined ..................... 101 4.20 Pearson Product-Moment Correlation for Total Internal (I) Scores with.Citizen Action Scores for each of the Sample Groups ..... . ............................ 103 4.21 Pearson Product-Moment Correlation for Total Chance (C) Scores with Citizen Action Scores for each of the Sample Groups ..................................... 104 v*ii Table 4.22 4.24 Pearson Product-Moment Correlation for Total Powerful Others (P) Scores with Citizen Action Scores with Citizen Action Scores for each of the Sample Groups ............................................... Pearson Product-MOment Correlation for Total Persuasive Action Scores with Citizen Action Scores for each of the Sample Groups ....................... Pearson Product-Moment Correlation for Total Political Action Scores with Citizen Action Scores for each of the Sample Groups ....................... Pearson Product-Moment Correlation for Total Ecomanagement Action Scores with Citizen Action Scores for each of the Sample Groups ................ Pearson Product-Mbment Correlation for Total Economic Action Scores with Citizen Action Scores for each Of the Sample Groups ...................... Pearson Product-Moment Correlation for Total Legal Action Scores with Citizen Action Scores for each of the Sample Groups ........................... Pearson Product-Moment Correlation for Total I, P, and C Scores within each Sample Group ............... Repeated Measures Analysis of Variance for I, P, C Scores by Sex for the Environmental Issues Class (Group I) ..................................... Repeated Measures Analysis of Variance for I, P, C Scores by Sex for the Biology Group (II A) Assigned the Solid Waste Issue ................... . ........... Repeated Measures Analysis of Variance for I, P, C Scores by Sex for the Biology Group (II B) Assigned the Acid Rain Issue ................................. Repeated Measures Analysis of Variance for I, P, C Scores by Sex for the Sierra Club Membership (Group III) .. ........... .. ....... . ............. .. Repeated Measures Analysis of Variance for I, P, C Scores by Sex for the Teachers Group Utilizing the Environmental Eduction Consultant (Group IV A) ....... viii 105 107 108 109 110 111 113 115 116 117 118 119 Table Page 4.34 Repeated Measures Analysis of Variance for I, P, C Scores by Sex for the Teachers Subgroup not Utilizing the District Environmental Education Consultant (Group IV B) ................................ 120 4.35 Repeated Measures Analysis of Variance for Total Action Category Scores by Sex for the Environmental Issues Class (Group I) ............... ....... ........... 123 4.36 Repeated Measures Analysis of Variance for Total Action Category Scores by Sex for the Biology Class (Group II A) Assigned the Solid Waste Issue ........... 124 4.37 Repeated Measures Analysis of Variance for the 'Total Action Category Scores by Sex for the Biology Class Assigned ‘the Acid Rain Issue (Group II B) ............. 125 4.38 Repeated Measures Analysis of Variance for Total Action Category Scores by Sex for the Sierra Club Membership (Group III) ................. .... ........... 126 4.39 Repeated Measures Analysis of Variance for Total Action Category Scores by Sex for the Teacher Group Utilizing the Environmental Education Consultant (Group IV A) ..... , .......... . ........ ...... ............ 127 4.40 Repeated Measures Analysis of Variance for Total Action Category Scores by Sex for the Teacher Group not Utilizing the Environmental Education Consultant (Group IV B) .............. ........... ..... ............ 128 4.41 One-way Analysis of Variance Between Biology Sub- groups II A and II B with Respect to Total PECM Scores ........ ........................................ 130 4.42 Repeated Measures Analysis of Variance with Respect to I, P, C Scores Between the Two Biology Sub- groups (Subgroups II A and II B) ..... ................. 131 4.43 One-way Analysis of Variance Between Biology Sub- groups II A and II B with Respect to Citizen Action Scores .......... . ........... ...... ............. 132 4.44 Analysis of Variance (Planned Contrasts) Based on Total PECM Scores Between Sample Groups Administered the Solid Waste Issue Summary ......................... 134 Table Page 4.45 Repeated Measures Analysis of Variance with Respect to I, P, C Scores Between the Two Teacher Subgroups (Subgroups IV A and IV B) ........... 135 4.46 Analysis of Variance (Planned Contrasts) Based on Citizen Action Scores Between Sample Groups Administered the Solid Waste Issue Summary ............ 136 LIST OF FIGURES Figure Page 2.1 Formula and Diagrammatic Representation of Rotter's Social Learning Theory .......................... ....... 20 2.2 Conceptual Summary of the Locus of Control Construct ... 21 2.3 Anatomy of an Environmental Behavior .... ............ ... 38 3.1 Examples of Internal (I), Powerful Others (P) and Chance (C) Statements as they Relate to Political Action ............................. .................... 48 3.2 Listing of the Five Instruments Used in Pilot Study I ........ ....................... ........... ...... 62 5.1 Pattern of I, P, C Scores for Group I and II B on the Internal/External Continuum ....... ... .. ..... .....149 5.2 Pattern of I, P, C Scores for Groups III and IV A on the Internal/External Continuum . ...... . ............ ....l49 5.3 Pattern of I, P, C Scores for Group IV B on the Internal/External Continuum . ........ ........... ...... ..149 5.4 Pattern of I, P, C Scores for Group II B on the Internal/External Continuum ....... . ..... .. ....... . ..... 151 xi Chapter 1 INTRODUCTION The Global 2000 Report to the President Major Findings and Conclusions If present trends continue, the world in 2000 will be more crowded, more polluted, less stable ecologically, and more vulnerable to disruption than the world we live in now. Serious stresses involving population, resources, and environment are clearly Visible ahead. Despite greater material output, the world's people will be poorer in may ways than they are today. For hundreds of millions of the desperately poor, the outlook for food and other necessities of life will be no better. For many it will be worse. Barring revolutionary advances in technology, life for most people on earth will be more precarious in 2000 than it is now — unless the nations of the world act decisively to alter current trends. The above excerpt presents just a sample of the major findings and conclusions published in a study prepared jointly by the State Department and the President's Council on Environmental Quality. This study was requested by the Carter Administration in 1977 and was completed and presented to the President in late 1980. A major objective of the study was to project current trends in world population, resources, and environmental change to the end of the century. The gloom and doom conclusions of the Global 2000 Report are by no means unique in general content. They only echo similar 1 % projections and predictions of many other reputable books and re- search studies published over the last two decades. An obvious question that plagues the world as a result of pro- jections like those in the Global 2000 Report is: How can nations, states, cities or individuals "act decisively" to prevent a collision course with these potentially awesome developments? The answer to a question as immense as this is not simple. However, although the answer may not be simple, there are many within the profession of Environmental Education (EE) who believe there is an answer. One seemingly idealistic, but in reality reasonable, solution that has been proposed is that educators around the world must begin to develop an "Environmentally Literate Citizenry." That is, a citizenry capable of identifying and investigating environmental issues and ultimately willing and able to take responsible action toward the remediation of those issues (Hungerford and Peyton, 1976). When major environmental issues are put in their proper per- spective (i.e., relative to human survival) environmental literacy seems like more than an appropriate goal for educators and citizens to strive for. A major component of environmental literacy is the willingness and ability of individuals to take responsible action on environmental issues (Peyton, 1977). This need to develop individuals who are responsible action—takers has been endorsed by the writings of many EE professionals (e.g., Hawkins and Vinton, 1973; Rilo, 1974; Loret, 1974; Hungerford and Peyton, 1976: Hungerford, et al., 1980; Childress, 1976; Belgrade Charter, 1976; Tbilisi Conference, 1978; Stapp and Cox, 1979). In order to achieve literacy as a major goal of EE it is necessary for educators to identify those factors within individuals which promote willingness and ability to implement environmentally responsible action. A general goal of this study was to develop a research instrument which could be utilized to increase effectiveness in promoting environmental action taking behaviors. Environmental education programming efforts aimed at achieving environmental literacy are indeed being pursued. However, many of these efforts have been based on the seemingly false assumption that a linear or domino type relationship exists among knowledge (cognitive), affect (attitudes) and behavior (conative) domains (Peyton and Miller, 1980; Burrus-Bammel, 1978). This relationship implies that a change in knowledge or beliefs influences attitudes which, in turn, will have an effect on behavior. Thus, many EE program objectives primarily address only knowledge and/or attitude development (Childress, 1978). It seems to be taken for granted that the desired behavior will follow. Research studies discredit the assumption of a linear relation- ship between knowledge, attitudes and behavior (Borden and Schettino, 1979; Burrus-Bammel, 1978; Heberlein, 1973; Ramsey and Rickson, 1977; Bowes et al., 1978 i3 Shoenfeld, 1980). In fact, these studies seem to indicate that the interacting variables affecting an indivi- dual's environmental actions are more complex than previously assumed and little understood (Peyton and Miller, 1980). Evidence does seem to exist which supports the premise that environmental action—taking is situation specific (Bowman, 1977; Stamm et al., 1977; Koenig, 1975; Sharma et al., 1977; Trichenor et al., 1973 ip_Shoenfeld, 1980). Thus, in any given person the type or number of variables affecting action taking may vary from situation to situation. Although it is not clear what variables affect action taking from situation to situation researchers have identified categories or types of environmental actions that people can or are prone to take across situations (i.e., environmental issues). Hungerford and Peyton (1980) present a three-part paradigm which identifies and de- ‘ fines specific categories of action, levels at which these actions i can be taken and finally, the constraints that may affect the taking of a particular action. Of particular concern to this study are the specific categories of environmental action; these are listed and defined below. Categories of Environmental Action 1) Persuasion: An effort to verbally motivate human being to take positive environmental action as a function of modified values, e.g., argumentation, debate, speech making, letter writing. 2) Consumerism: An economic threat by an individual or a group aimed at some form of behavior modi- fication in business or industry (e.g., boycotting) or some conservative mode of behavior with respect to goods and/or services (e.g., discriminating and conservative use of goods and services). 3) Political Action: An effort aimed at persuading an electorate, a legislator (or legislature), or executive governmental agency to conform to the values held by the person or persons taking that action, e.g., lobbying, voting, supporting candidates. 4) Legal action: Any legal/judiciary action taken by an individual and/or organization which is aimed at some aspect of environmental law enforcement or, a legal restraint preceding some environmental be— havior perceived as undesirable, e.g.,lawsuits, injunctions. 5) Ecomanagement: Any physical action taken by an individual or a group aimed directly at main- taining or improving the existing ecosystem, e.g., reforestation, landscaping, installing bird boxes. In a final analysis, although investigators have an idea of what environmental actions can be or are taken, it would appear that findings are inconclusive as to how or what attitude and knowledge variables interact to promote or extinguish an individual's action taking behavior. Clearly more research into the causes of environmental action taking behavior is warranted if educators hope to enhance the development of environmentally literate citizens. Locus of Control and Environmental Action A specific attitudinal variable which may impinge on the en- vironmental action taking behavior of an individual is his/her perception of personal control over a situation or event (e.g., environmental issue). This perceived belief about personal control or non-control of an event is directly related to the theoretical construct called Locus of Control (L of C). The L of C construct represents one of four equally weighted components of Rotters'(1966) Social Learning Theory (SLT). In its most basic form, the SLT states that "the potential for behavior to occur in any specific psychological situation is a function of the expectancy that the behavior will lead to a particular reinforcement in that situation and the value of that reinforcement." (Rotter, 1975, pp. 57). The four components of the SLT are: behaviors, ex- pectancies for reinforcement, value of reinforcement, and the psychological situation. A formula and diagrammatic representation of SLT are presented in Chapter 2 of this study. It is the expectancy for reinforcement, or the "probability held by the individual that a particular reinforcement will occur as a function of a specific behavior on his part in a specific situation or situations" (Rotter, 1954, pp. 107) which constitutes an individual's L of C. Phares (1976, pp. 16) describes the probability held for reinforcement (i.e., expectancy for reinforcement or L of C) as a "subjective probability... It is determined not just by one's Objective past history of reinforcement" (i.e., specific expectancy) ” but also by expectancies generalized from other, related behavior-reinforcement sequences" (PP. 16). Thus, when individuals are in a relatively unfamiliar situation, generalized expectancies will be relied upon: When individuals have a great deal of experience in a given situation, specific expectancies will be heavily relied upon and generalized ex- pectancies will have less significance. Rotter has identified two belief positions an individual may harbor with respect to specific or generalized expectancies Of re- inforcement (L of C). When a reinforcement is perceived by the subject as following some action of his own but not being entirely contingent upon his action, then, in our culture, it is typically perceived as the result of luck, chance, fate, as under the control of powerful others, or as unpredictable because of the great complexity of the forces surrounding him... We have labeled this belief in external control. If the person perceives that the event is contingent upon his own behavior or his own relatively permanent characteristics, we have termed this a belief in internal control (Rotter, 1966, p. 1) (emphasis added). Given that environmental action may be "situational" and that there are specific actions that can be applied across situations/ environmental issues, it is possible that an individual's willingness and ability to take action will, in part, be influenced by his expectancy (L of C) that the action taken will indeed produce the desired (i.e., valued) outcome or reinforcement. If one's expectancy is high for obtaining the desired results through use of some action, then the person could be described as internal about taking that action in that situation. If one's expectancy is low for Obtaining the desired results through use Of some action, then the person could be described as external about taking that action in that situation. It might also be hypothesized that a person who is internal about an action will be more likely to take that action than a person who is external about it. Measuring L of C Since its initial introduction, well over 600 research studies have concentrated on the L of C construct and this number doesn't even take into account the unpublished master's theses and doctoral dissertations (Rotter, 1975). Many of these studies included the development of instruments designed to measure an individual's L of C. Using these instruments both generalized and specific expectancies have been assessed in a variety of age groups, populations, and content areas. Phares (1976) and Lefcourt (1976) both present a sampling of the different types Of instruments that have been developed and tested. Probably the most widely tested L Of C instrument that has been developed is Rotters' (1966) Internal-External (I-E) Scale (Appendix A). This scale consists of 23 forced-choice items which are designed to measure expectancies across a wide range of situations, such as interpersonal situations, school, government, work and politics (Phares, 1976, p. 42). Because the I-E scale is designed to measure expectancies across a variety of areas it is labeled a generalized expectancy scale. In contrast, if the scale is designed to measure one content area (e.g., environmental action taking) then the scale is described as a situation specific scale. Rotter (1975), Lefcourt (1976), and Phares (1976) all indicate that a measure of broad generalized expectancy allows for predictions in a variety of situations, but it may function at a low level in trying to predict actions in a specific situation. Therefore, the I-E scale, a measure of generalized expectancy, may do a good job of predicting people's behavior in general but it may miss rather badly in specific situations (Phares, 1976). On the other hand, a narrower, more situation specific measure should allow relatively high levels of prediction in the situation it was designed for but it is rather limited in breadth of application. If the only purpose is to explore environmental action taking behavior, then it might prove advantageous to construct an I—E scale with items that pertain solely to environ- mental action taking behavior. In an attempt to refine the predictability of the I-E scale, instruments have been developed which distinguish between different types of externals (Crandall et al., 1965; Kleiber et al., 1973; Levenson, 1972). Hanna Levenson (1972) constructed three generalized expectancy scales consisting respectively of Internal, Powerful Others, and Chance items (IPC Scale). She felt that peOple who believe the world is unordered (i.e., chance oriented) should be- have differently from those who feel powerful others are in control; although both chance and powerful others would be considered external beliefs. In some situations Levenson's (1972a, b; 1973a, b, c; 1974) approach has shown a fair amount of success in discriminating between individuals with these three belief systems. This tripartite analysis of Levenson's could be particularly applicable to exploring environmental action taking behavior. It is possible there are important behavioral differences between the external who feels that the outcome of a certain environmental issue is unpredictable (i.e., chance oriented) and the external who perceives the outcome is predictable but powerful others are in control. Although.Levenson's IPC scale is more sensitive to different external orientations than the I-E scale, it is still a measure of generalized expectancy. Based on this similarity, the IPC scale, like the I-E scale, could be subject to the weakness of low level behavior prediction in specific situations. If the goal of a L of C instrument is to analyze certain be- haviors in one situation or a somewhat homogeneous group of situations, then it might prove worthwhile to develop a scale with all items directed towards behavior in that situation(s) (Phares, 1976). Studies using generalized instruments have shown some indication of a relationship between L of C and environmentally responsible 10 behavior (Levenson, 1972; Tucker, 1978; Arbuthnot, 1977; Smith, 1979). It seems highly possible that these relationships could be better explored with a situation specific instrument. Although the need seems apparent (Peyton and Miller, 1980; Smith, 1979; Tucker, 1979), to date no situation specific L of C instrument has been developed to explore environmental action taking behavior. Statement of the Problem The purpose of this research study wastp develop a situation Specific Locus of Control that can be used to measure an individual's perceived expectancy for environmental action taking in a stated situation. Specifically, the instrument assesses three L of C belief systems that could be applied by an individual to a stated situation (i.e., specific environmental issues). First, the in- strument was designed to measure the degree to which an individual perceives himself/herself as in control of affecting the outcome of a stated situation through use of a) persuasive actions; b) economanagement actions; c) economic actions; d) legal actions; e) political actions. Second, it was designed to measure the degree to which an individual perceives powerful others to be in control of the stated situation regardless of the environmental actions he/she may take. Third, it was designed to measure the degree to which an individual perceives chance or fate to be in control of the stated situation regardless of the environmental actions he/she may take. 11 Significance of the Study It is generally agreed upon by the EB community that its ultimate goal is to develop a citizenry that is both able and willing to take responsible action toward the remediation of environmental issues. To pursue this goal it would be advantageous for educators to better understand the interacting variables that impinge upon an individual's willingness and ability to take responsible actions. In this study L of C is presented as one variable which may play a role in affecting the environmental action taking behavior of an individual. Since its inception, a plethora of investigations have centered on the L of C construct and, as a result, several generalizations about the behavior of individuals can be inferred (Phares, 1976; Lefcourt, 1976). Peyton and Miller (1980) identified and presented the following L of C generalizations and their implications for BB. 1. Internals more frequently participate in productive action takingpthan externals. ... The relationship between internality and individual action taking has strong implications for EE. Achieving the goals of EB depends on developing individuals willing to initiate positive, rational environmental action taking. In view of the generalizations reported here, this would seem to make internality desirable. 2. Internals differ from externals in their ability to recall relevant material, and in how actively they seek additional information. ... Having greater recall of relevant material and more actively seeking additional information are certainly important abilities for effective en- vironmental problem solving. If it is accurate that becoming more internal leads to increases in the above characteristics (causal relationship), then developing an internal L of C among citizens may be an important goal of EE. 3. Internal individuals are superior to externals in their utilization of information. ... Rational, objective problem solving would be en— hanced by an increased ability to accurately apply information. If the relationship between this characteristic and internality is a causal one (i.e., becoming more internal would cause a greater utilization of information ), citizens' perception of L of C should be an important consideration of environmental educators. 4. Internal individuals are more resistant to subtle manipulation and are less influenced by high—prestige individuals than externals. ... It is essential that the value positions and credi— bility of informational sources be carefully assessed when investigating the dimensions of an environmental issue. It seems reasonable to expect internals to be more capable and/or willing to reject information which comes from biased or prestigious, but uninformed sources. 5. Internal individuals exhibit a superior capacigy to delay gratification in order to attain greater, long- term gains. ... Solving environmental (and other social) problems, often requires behaviors that sacrifice short—term rewards for the attainment of greater, long—term gains. If EE is to produce citizens capable and willing to adopt behaviors to improve and/or maintain environmental quality, increased internality may be an important part of the process. -——4 l3 6. Internals respond differently to those tasks which they perceive to be skill-related, than to tasks they perceive to be chance-related. ... In view of the above findings, it appears im- portant for BB to present citizens with the perception that the outcomes of environmental actions are skill- related and not due entirely to chance events. 7. An individuals' perceived L of C is susceptible to change. ... Given that an internal L of C in citizens is accepted as a desired perspective in an environ- mentally literate individual, it is significant that L of C is responsive to training and experience. The nature and extent of such training to be offered by EB are by no means clear yet. However, the implications seem evident that environ- mental educators should begin to examine EE curricula and teaching methods to determine how an internal L of C may be best developed in citizens. Although the inferred relationships between environmental action and L of C seem quite apparent, there has been little effort to investigate their authenticity. Those studies that do provide some indication of a linkage between L of C and environmental action are based on generalized instruments. Major proponents of L of C theory state that if the intent of a study is to measure behavior in a homogeneous class of situations then it would be beneficial to con- struct a scale with all items directed toward that particular type of situation. To date, no situation specific instrument has been developed which can be used to investigate the inferred relation- ships between environmental action taking behavior and L of C. The development of such an instrument was the proposed topic of this study. 14 Limitations of the Study Two Specific categories of limitations can be identified in this study: (1) Limitations imposed by population variables; (2) Limitations imposed by instrument design. Population Limitations The sample size used and characteristic homogenity of the populations tested was determined by the availability of participants. Due to lack of random sampling, generalizations from results to other non-tested populations should be made with caution. Limitations of Instrument as Designed The instrument was designed to measure an individual's perceived expectancy for personal use of environmental actions. It is possible that a person's perceived expectancy will not re- flect actual behavior. The respondant was placed in a hypothetical situation and asked to react to that situation. ReSponses could have been dif- ferent if it were a real-life situation. The instrument was designed as a self-report measure. Under certain conditions, the individual's belief system may be in conflict with social norms and the respondant may attempt mahidé true beliefs when reSponding. Strong evidence of instrument validity can only be achieved through.continued application of an instrument. This study only attempts to initiate evidence of instrument validity. 15 5. The instrument designed in this study was not tested against a generalized L of C instrument to determine its comparative diagnostic potential. Goals and Strategies of the Study The general goal of this research study was to construct a situation specific L of C instrument which could be used to measure the perceived expectancy of an individual for taking specific types of environmental action in a given situation. In addition, this instrument was tested for initial evidence of validity and reliability. For purposes of clarification, the instrument designed in this study was entitled the Perceived Environmental Control Measure (PECM). To achieve the above goal a series of strategies were considered which 1) established test item content; and, 2) tested for initial evidence of reliability and validity of the instrument. Strategies for Establishingpitem Content An item pool was deveIOped with statements divided equally among three belief orientations: 1) Internal (I); 2) Powerful Others CP); 3) Chance (C). Furthermore, these belief systems were applied equally across five (5) categories of environmental action: 1) Legal Action; 2) Persuasive Action; 3) Political Action; 4) Ecomanagement; 5) Economic Action. 1. Items in the I-scale were constructed to elicit re- Sponses which measure the degree to which an individual ...perceives that his/her use of an environmental action will have an effect on or control the outcome of a stated situation. 16 Items in the P-scale were constructed to elicit re- sponses which measure the degree to which an individual ... perceives powerful others, more than his/her own use of an environmental actiOn, will control or have an effect on the outcome of a stated situation. Items in the C-scale were constructed to elicit re- sponses which measure the degree to which an.individual ... perceives chance or fate, more than his/her use of an environmental action, will control or have an effect on the outcome of a stated situation. Strategies for Testing of the Instrument In order to gain initial evidence of reliability and validity, it was proposed that two pilot studies and four field testings of the instrument be conducted. Strategies used to accomplish each of these studies are presented below. Pilot Study I 1) administer five (5) separate instruments with a combined 2) 1) total item pool of 150 statements. i. 30 statements IPC Ecmmanagement Instrument ii. 30 statements IPC Persuasive Action Instrument iii. 30 statements IPC Economic Action Instrument iv. 30 statements IPC Legal Action Instrument v. 30 statements IPC Political Action Instrument analyze and select items for second pilot study based on item total correlation, item subscale correlation, item response distribution, and mean responses. Pilot Study II collapse qualifying items from pilot study one into one instrument. Administer that instrument along with a citizen action questionnaire. (Theoretically, internals should be more involved than externals in taking responsible citizen actions.) 2) Analyze and select items for the final instrument based on relatively high item total correlations, relatively high item subscale correlations, item response dis— tribution and mean responses. Field Testing of Instrument 1) Administer a citizen action questionnaire along with a final version of the Perceived Environmental Control Measure as developed from pilot studies one and two. 2) Analyze the instrument for evidence of subscale re— liability, whole test reliability, content validity, and construct validity. 3) Investigate the following general research questions: 1. Can evidence of validity and reliability be established for an instrument developed to measure perceived L of C toward taking Specific categories of environmental action in a stated situation? 2. Do selected sample groups differ significantly on Perceived Environmental Control Measure scores? b.) What relationship exists between scores on a measure of citizen action and Perceived Environ- mental Control Measure scores for a selected group of respondents? b Do total Internal (I), Powerful Others (P) and Chance (C) scores from the Perceived Environ- mental Control Measure differ significantly for a selected group of respondents? U1 0 Do total (i.e., combined ) I + P + C scores for each environmental action category in the Perceived Environmental Control Measure differ significantly for a selected group of respondents? 6. Are the Perceived Environmental Control Measure Scores of a selected group significantly affected by different environmental issues summaries? * Note: These research questions are more specifically defined and/or expanded in Chapter III of this study. Chapter 2 REVIEW OF RELATED LITERATURE "The success of public policy decisions,educational programs and other efforts dependent upon specific individual action in the realm of environmental issues may well hinge upon our understanding of the re- lationships among personality characteristics, attitudes and environmental values, knowledge and behavior" (Arbuthnot 1977, pp. 217). Locus of Control (L of C) is proposed as one of several interacting variables which affect the environmental action taking behavior of an individual. This chapter will present a discussion of: (1) the theo- retical origin of L of C; (2) techniques used to measure L of C; (3) categories of environmental action taking behavior; and (4) studies in- dicating a relationship between environmental action taking behavior and L of C. Theoretical Origin of Locus of Control The nature of Locus of Control (L of C) can best be understood through an examination of its theoretical origin. Locus of Control was first introduced as a component of J. B. Rotter's Social Learning Theory (Rotter 1954; Rotter, Chance, Phares 1972). This theory was first developed in an effort to explain and investigate the social behavior of psychotherapy patients. Since the introduction of Rotter's Social 18 19 Learning Theory (SLT) considerable empirical evidence has been established which supports its utility for explaining human behavior both in and outside of the clinical setting. Rotter's SLT states that a person's actions or behaviors are a function of three equally interacting components: expectancy for re- inforcement, value of reinforcement and the psychological situation. A "reinforcement" can be described as "anything that has an effect on the occurrence, direction, or kind of behavior: (Phares 1976, pp. 15). The "value of a reinforcement" may be defined as "the degree of preference for any reinforcement to occur if the possibilities of their occurring were all equal" (Rotter 1954, pp. 107). The "expectancy for reinforcement" is the ' 'probability held by the individual that a particular reinforcement will occur as a function of a specific behavior on his part in a specific situation or situations" (Rotter 1954, pp. 107). The psychological situation is the accumulation of cues that might directly affect the expectancies and reinforcement values of a given person (Phares 1976, pp. 17). Figure 2.1 presents a formula and diagrammatic repre— sentation of SLT (adapted from Lefcourt 1976, pp. 26). It is the expectancy for reinforcement which reflects an indivi- dual's L of C. The model in Figure 2.2 presents a conceptual summary of the L of C construct. Components of the model are explained below. Expectancies fall into two categories: specific and generalized. Generalized expectancies are those which originate or are drawn from a variety of life's experiences. Specific expectancies refers to those which originate from a particular experience or homogeneous class of experiences. When an individual is in a novel or unfamiliar 20 and RV ) a BPx,isa = (Ex,Rasl $1 The potential for behavior (x) to occur in situation (7) in relation to reinforcement (a) is a function of the expectancy of the occurrence of reinforcement (a) following behavior (x) in situation (7) and the value of reinforcement (a) in situation (1) Figure 2.1 Formula and Diagrammatic Representation of Rotter's Social Learning Theory (adapted from Lefcourt 1976, pp. 26) 21 EXPECTANCY "Probability held by an individual that a particular reinforcement will occur as a function of a specific behavior on his part in a specific situation." v LOCUS OF CONTROL L of C Shown to operate as generalized expectancy and specific expectancy "Continuum" "EXTERNAL CONTROL" "INTERNAL CONTROL" ...perceives that the ...perceives that the result of an event is result of an event is contingent upon his own contingent upon luck, behavior or characteristics. chance, fate or powerful others... Control Construct Figure 2.2. Conceptual Summary of the Locus of - 22 situation generalized expectancies will be important in determining expectancy for that situation. If an individual has a great deal of experience in a given situation then specific expectancies will be the primary determinants (Phares 1976). Rotter (1966) has further identified two belief systems an in- dividual may develop as a result of generalized and/or specific expec- tancies for reinforcement. When a reinforcement is perceived by the subject as following some action of his own but not being entirely contingent upon his action, then, in our culture, it is typically perceived as the result of luck, chance, fate, as under the control of powerful others, or as unpredictable because of the great complexity of the forces surrounding him... We have labeled this belief in external control. If the person perceives that the event is contingent upon his own behavior or his own relatively permanent characteristics, we have termed this a belief in internal control (Rotter 1966, p. 1). (emphasis added) The L of‘C construct is also described by Rotter (1966) as a dis— tribution of individuals on a continuum according to the degree to which they accept personal responsibility for what happens to them. Lefcourt (1976) states that people are not totally internal nor external. The terms are not meant to imply that perception of control is a trait or typology. The terms internal and external control depict an individual's more common tendencies to expect certain events to be contingent or non—contingent upon their action. Measuring Locus of Control Lefcourt (1976) states that given the current status of assessment tools used in the study of L of C it would be possible to conclude 23 that there is enough evidence to encourage investigators to both continue in their use of existing devices and to develop more criterion specific measures. A review of the research (Lefcourt 1976, Phares 1976) indicates that both these avenues of assessment have been pursued. The most widely tested instrument is Rotter's Internal — External Scale (I-E Scale, Appendix A). This instrument consists of 23 forced- choice items which attempt to sample beliefs across a wide range of situations such as school, government, work, politics and interpersonal situations (Phares 1976). The I—E scale has been very successful in establishing the validity of the L of C construct but since its de- velopment considerable efforts have been expended to improve and refine L of C measurement. Two areas of improvement or refinement of L of C measurement that are pertinent to this study are (1) development of situation specific instruments; and (2) discrimination between different types of externals. Generalized vs Situation Specific Measures of Locus of Control Because the I-E scale was developed to sample beliefs across a wide range of situations it represents a generalized expectancy scale. In contrast a scale designed to measure one content area such as environ- mental action taking would function as a situation specific scale (Phares 1976). Major pr0ponents of the L of C construct state that measures of generalized expectancy allow for predictions in a large number of dif- ferent Situations, but at a low level (Lefcourt 1976; Phares 1976; Rotter 1975). A situation specific measure allows more accurate prediction 24 in the situation it was designed for, but its utility is confined to that specific type of situation. Thus, if the goal of a researcher is to predict behavior in one situation or a very homogeneous class of situations, then it would be advantageous to develop a scale with all items directed towards that situation. If the intent is to sample L of C beliefs across a wide range of situations (i.e., life in general) then a generalized scale would be warranted (Phares 1976). A number of research studies have been conducted which support the premise that L of C instruments designed for specific situations can yield more refined predictions than a generalized scale applied to that same situation. Abramowitz (1973, in Phares 1976) compared college students having social action or political goals with those who were not political in nature nor social action-oriented. He obtained three types of I-E scores; one based on all 23 items, one based on world events or politically-worded items and a third based on non-political or personal control items. It was found that neither the total I-E scores nor the non—political item scores related to sociopolitical behavior. In contrast, a positive relationship between internality on the political items and action—oriented sociopolitical behavior did exist. In another study Donovan and O'Leary (1978) present data which indicate that a situation Specific instrument designed to address drinking related behavior differentiated Significantly between alcoholics and non—alcoholics. It was reported the same populations did not differ with reSpect to scores on Rotter's measure of generalized expectancy (I-E scale). 25 Further support for the utility of designing situation specific instruments is provided by Wallston et al. (1976). These researchers assumed that a specially constructed health related L of C scale would provide more sensitive predictions of the relationship between internality and health behaviors than would the I—E scale. Eighty-eight college students were given the Health L of C scale, the I-E scale, and a measure of the relative value they placed on health. Subjects next read a ”mildly-threatening" message about the dangers of hypertension. In addition, they took a difficult knowledge test on hypertension which was designed to reinforce the feeling that this was a subject about which they knew little. After the message—test treatment, students were given the oppor— tunity to broaden their knowledge on hypertension by choosing to read any number of 16 different information pamphlets on the topic. It was proposed that subjects who held internal L of C beliefs and who highly valued health would choose to expose themselves to more information (i.e., more hypertension pamphlets) about the given condition. Results indicated that there was a "marginally significant" in- teraction between perceived health value and Health L of C scores. No such trend was found between I—E scale scores and perceived health value. In addition, when Health L of C scores were used as the basis of classification, high health value internals chose more pamphlets than all other types of subjects. The researchers state that no such re- sults would have been evident if the more general I-E scale had been the only basis for classifying subjects as internals or externals. 26 Wallston et al. (1976) also conducted a second study on weight loss behavior of 22 overweight women. It was hypothesized that subjects in a weight reduction program whose orientation was consistent with their expectancies (L of C) would be more satisfied and more successful than subjects in a program inconsistent with their L of C beliefs. Prior to treatment, subjects were given the Health L of C scale and the I-E scale. Subjects matched on Health L of C scale were randomly assigned to one of two weight reduction treatments: an internally oriented self—directed program, or an externally oriented group program. After an 8-week treatment period data on program satisfaction and weight loss were collected. Results indicated a significant interaction between Health L of C scores and program satisfaction. Those programs consistent with subjects' expectancies were evaluated more positively than were inconsistent programs. That is, internals in the self-directed program were more satisfied than those in the group program. Externals in the group program were more satisfied than those in the self—directed program. Internal-External Scale scores indicate those in expectancy consistent programs did not differ in perceived satisfaction from those in programs inconsistent with expectancies. Data on weight loss was less clear. Statistically significant results were not reported but the researcher claim results were in the expected theoretical direction when applying Health L of C scores. Externals in the group program lost more weight than did those in the self-directed program. Internals in the self-directed program lost more weight than did those in the group program. Internal-External scale scores indicated weight loss results were in a direction opposite to theoretical prediction. 27 Given the findings presented above it would seem that generalized L of C instruments may have limited usefulness in specific Situations. Locus of Control scales with all items directed at the situation to be studied appear to be more effective in discovering L of C relationships. An appropriate application of this type of instrument may be in the area of specific categories of environmental action as proposed by Hungerford and Peyton (1980). Developing a L of C instrument with items directed at these actions may prove to be a better predictor of environmental action taking behavior than a generalized instrument. Agents of External Control A second means of refining I—E measurement involves the issue of unidimensional vs multidimensional control (Lefcourt 1976). Rotter's I-E scale is described as being unidimensional. It measures the degree to which people believe they exercise control over their lives (internal orientation) or the degree to which they feel their destinies are beyond their own control and are determined by fate, chance, or powerful others (external orientation). Several investigators have found reason to suggest that I-E functions as a multidimensional rather than a unidimensional construct (Crandall et a1. 1965; Hersch and Scheibe 1967; Gruinet a1. 1969; Mirels 1970; Levenson 1972a; Sanger and Alker 1972, in Reid and Ware, 1974). For example, a person who believes powerful others control events may behave differently from a person who is chance oriented. However, both chance and powerful other belief ori- entations are groups under the rubic of external control in the unidimensional I-E theory. 28 Phares (1976) points out that a multidimensional concept is apparent in Rotter's (1966) original definition of I-E. Although many expec- tancies are considered equivalent (i.e., chance orientation = powerful others orientation = lack of freedom, etc.) they may on occasion lead to dissimilar behaviors. Thus, there is understandable evidence of the existence of multidimensional factors in I—E construct but there is much less evidence that demonstrates the predictive utility of these factors. Phares (1976) has called for the development and testing of these factors. Hannah Levenson has probably exerted the most effort toward ex— amining the utility of assessing diverse agents of control (Lefcourt 1976). Levenson (1972b) proposed to measure three separate L of C belief orientations: Internal (I), Powerful Others (P) and Chance (C). Powerful others and Chance orientations are presented as two independent external belief systems. The rational for factoring out these two types of externals was that people who believe the world is unordered (chance-oriented) would behave and think differently from people who believe the world is ordered but powerful others are in control. In the latter case, a potential for control exists (Levenson 1974). With a fair amount of success, Levenson (1973b) conducted a number of studies which serve to ascertain the validity of separating L of C measurement into I, P, and C dimensions.' The three independent orientations were found to emerge in a series of factor analyses and were shown to be differentially related to such variables as philosophy of human nature, involvement, information, activism and perceived parental upbringing. Some examples of validating studies are discussed below. 29 Factor Analysis of the I, P, and C Scales Using reSponses from a population of college males (N=239) and a population of psychiatric patients (N=165), two factor analyses were computed. Results were as follows: For each of the samples, the first three factors to emerge were I, P, and C accounting for 60% of the total variance. Both analyses approach the ideal simple structure, Since there is almost pp overlap of the items on the factors and each I, P, and C factor remains conceptually pure in that only items from the appropriate scale load on the one factor. The dimensions of internal control, powerful others, and chance seem to be consistent points of reference 1 for both normal and abnormal samples. (Levenson 1973 (b), pp. 3). Another study done by Walke (1979) seems to confirm Levenson's factor analysis findings. The I, P, C scales were administered to 71 male and 85 female New Zealand college students. A factor analysis of these responses according to Walke "clearly confirmed the three—factor structure underlying Levenson's questionnaire..." (pp. 532). Citizen Involvement and the IPC Scales Levenson (1972b) studied a population of 96 adults to determine if I, P, C scores could be related to antfpollution behaviors. One third of the SS were selected randomly from the membership of a local anti-pollution group. Another third were selected from those who knew of the anti-pollution but decided not to join. The final third were not aware of the anti-pollution group. A series of scales including the I, P, C scales, an Involvement Activities Checklist and a measure of knowledge about pollution were administered to all the subjects. Levenson (1973b, pp. 3) presents the following summary of results. 30 Results indicated that while the C scale had no relation- ship to involvement for females, males who believed that chance did not control their lives were involved in significantly more activities than those who per- ceived that chance had more control. No significant results relating the I and P scales to involvement were found. Similarly, male non-members scored significantly higher on the C scale than did male members. There were no significant differences be- tween members and non—members on the I or the P scales. It thus may be reasoned that expectations of control by powerful others or low expectations for control by self do not diminish involvement because the potentials for control still exist. For the high chance be- liever, however, there would be no such hope of control and so high C scale scores should be less involved. Information and IPC Scales Seman and Evans (1962) conducted a validation study of the I-E scale by relating scores to the amount of information hospital patients had with respect to their illness. The theoretical rationale behind the study was that people who were internally oriented would attempt to control their environment through knowledge (Levenson 1973b). Results indicated that the more externally oriented the patient, the lower he/ she scored on an objective test about their illness. A similar validation study was used with the I, P, C scales and knowledge of pollution matters. The sample population included subjects who were known to be members of the anti-pollution group discussed in the previous section of this paper (i.e., Citizen Involvement and I, P, C Scales). It was found that males who believed chance or fate controlled their lives had significantly less information than did those who felt chance did not control their lives. No significant relationship was found between P or I scores and amount of information. 31 Student Activism and I, P, C Scales Levenson (1973b) reports that a Master's Thesis done by Jim Miller at Texas A & M involved the administration of Kerpelman's Political Activity Scale - a measure of Conservationism—Liberalism and the I, P, and C scales to 99 undergraduate students. The intent of the study was to investigate possible relationships between student political ideology and activism and IPC orientation. Analysis indicated that the I scale scores were unrelated to ideology or activism. However, significant findings with the C scale indicate that conservatives believe less in control by chance forces than liberals. In reference to the P scale, liberal activists had signi- ficantly higher P scale scores than conservative activists. Levenson speculated that conservative students might be discouraged from activism if they have a high perception of powerful others, while such perceptions might encourage the aetivism of liberals, who are by definition against the status quo. Parental Anticedents and I2 P, C Scales In an effort to examine the relationship between parental rearing and expectancies of control the I, P, C scales and the Perceived Parenting Questionnaire were administered to 276 undergraduates (Levenson 1973a, 1973b). Parental behaviors associated with internality were perceived differentially depending upon sex of the respondent. Males who perceived themselves to be helped and taught by their mothers had higher I scale scores. There was no such findings for females. However, girls who perceived that their mother did not worry about them had 32 significantly higher internal scores than those who thought their mothers were protective. Individual's perceptions of the extent to which their parents were demanding, punishing, and controlling were all positively related to scores on the powerful others scale. Subjects who had problems in figuring out what to expect from their parents had significantly more perceptions of the world as unordered or as controlled by chance. In addition to the above validation studies, Levenson has also found IPC factors operating differentially among prison inmates (Levenson 1974) and among psychiatric patients (Levenson 1973c). In summary, it appears that the refinement of the I—E scale into the tri— partite I, P, C division is justified by a number of studies. Levenson has made a contribution to L of C measurement by demonstrating the utility of employing these factored scales. If, in fact, those who believe in powerful others differ from chance oriented people (as research seems to indicate) in cognitive and behavioral outcomes then it might prove advantageous to apply these two belief systems toward the expectancy assessment of environmental action taking behavior. Taking this direction may lead to a more sensitive analysis of proposed relationships be- tween L of C and environmental action, than if the unidimensional I-E approach were pursued. Environmental Action Categories Locus of Control theory contends that a person's expectancies may change from situation to situation. For example, an individual may 33 behave ina.predominantly internal fashion when dealing with academic goals but be significantly more external in his behavior when love and affection are the goals involved (Phares 1976, pp. 40). Given the variance in expectancy from situation to situation L of C prediction ought to be enhanced when we measure perceived L of C as it related to a specific situation. Environmental issues or problems present a rather diverse set of situations. Causes and effects will undoubtedly vary from issue to issue. Although the casues and effects may vary there are similar types of environmental actions that can be applied toward the remediation of these varying situations. Reference to specific types of actions that can be taken across situations is made by Stapp (1971): Specifically, citizens make these decisions as they cast votes on community issues; as they elect repre- sentatives to policy-making bodies; as they directly act upon the environment itself... They can ask informed questions... serve on advisory and policy-making committees... support sound legi— slation... (pp. 105-106) Hirst and Schuck (1971) list the following as possible actions that can be taken by an individual or group "(a) leading an ecologi- cally sensible life; (b) joining local, regional and national environ— mental organizations; (c) writing and influencing legislation and legislators; and (d) pressing legal actions against polluters." (pp. 204). Mason (1974, pp. 40—41) has classified the responses of society to an environmental crisis into five categories. Peyton (1977) States that these categories could be considered as a classification scheme for environmental actions. _¥— 34 (1) Educational - the creation of under-graduate and professional programs in environmental education at all levels of education, and government... and increasingly in those industries most closely associated with resources of the environment; (2) Political - the generation of a number of issues related to the environment that have varying degrees of voter interest in all levels of the political process; (3) Economic - the institution of various types of resource conservation measures including re- cycling, rationing, and special taxation designed to shift some of the costs of production from the environment back to the marketplace; (4) Legal — the apparent interest in environ— mental law and attorneys as advocates for the environment with resultant challenge to existing laws and redefinition of terms rele— vant to new legislation related to the environment; (5) Cultural - the apparent attempt to variously institutionalize a type of land or environmental code of ethics or standards within which society may choose to operate as it interacts with the environment in coming generations. Hungerford and Peyton (1980) have attempted to combine the various environmental actions identified in the literature into a three- part paradigm. They identify and define specific categories of action, levels at which these actions can be taken, and criteria for selecting a particular action. 1) Categories of Environmental Action Persuasion: An effort to verbally motivate human beings to take positive environmental action as a function of modified values, e.g., argumentation, debate, Speech making letter writing. 35 2) Consumerism: An economic threat by an individual or a group aimed at some form of behavioral modification in business or industry (e.g., boycotting) or some conservative mode of behavior with respect to goods and/or services (e.g., discriminating and conservative use of goods and services). 3) Political Action: An effort aimed at persuading an electorate, a legislator (or legislature), or executive governmental agency to conform to the values held by the person or persons taking that action, e.g., lobbying, voting, supporting candi- dates. 4) Legal Action: Any 1egal/judiciary action taken by an individual and/or organization which is aimed at some aspect of environmental law enforcement or, a legal restraint preceding some environmental behavior perceived as undesirable, e.g., law suits, injunctions. 5) Ecomanagement: Any physical action taken by an individual or a group aimed directly at main— taining or improving the existing ecosystems, e.g., reforestation, landscaping, installing bird boxes. Indirect economic actions of an individual are not well represented in the environmental action categories proposed by Hungerford and Peyton. For example, it could be argued that membership in environ- mental organizations and/or donations to environmental causes con— stitutes environmentally responsible behavior. Although these may be indirect environmental actions, they nonetheless involve more than a verbal committment by the contributing individual. This investigator would propose that the title of Consumerism be changed to Economic Action and redefined as: Economic Action: Constitutes an action similar to one of the following; a) an economic threat by an individual or group aimed at some form of be— havior modification in business or industry, e.g, boycotting; b) some conservative mode of behavior 36 with respect to consumption of goods and services, e.g., purchase of recycled materials; c) some monetary contribution to an individual, group, or institution that actively favors or works for a position supported by the contributor, e.g., donations to environmental causes; membership fees paid to environmental activist organizations. Economic action, as defined above, provides a more comprehensive approach to the various monetary actions that can be brought to bear by an individual on an environmental issue. Relationship Between Environmental Action Taking and Locus of Control Locus of Control has been assessed across a variety of age groups, populations and topic areas. Results from .these investigations allow a number of well grounded generalizations to be made about L of C's relation to human behavior (Lefcourt 1976; Phares 1976). Peyton and Miller (1980) have described several L of C generali- zations, supporting research and implications for EE. These were summarized in Chapter one of this study. In another section of the Peyton and Miller paper a model is presented in an attempt to graphically relate some of the many variables and processes which effect environ- mental action taking behavior (Figure 2.3). Within this model the conceptual niche of L of C, as an impinging variable on environmental action taking, can be visualized. The model analyzes the behavior of writing a letter to a senator about the Alaskan Land Bill, and is generally explained as follows: The left side of the model includes those impinging variables (knowledge and experiences), beliefs (B1 and B2), and attitudes (Al), which EE has tradi- tionally dealt with. More recently EE literature has encouraged, including a knowledge of, and 37 Figure 2.3. Anatomy of an Environmental Behavior (Peyton and Miller, 1980, pp. 184). ANATOMY OF AN ENVIRONMENTAL BEHAVIOR Knowledge of environ. Knowledge of action skills; case Alaskan lands Knowledge of natural systems studies of effective issue (ecology; vulnerability; etc.) action // I M P I N G I N G V A R I A B L E S (new knowledge and experiences) Affective Knowledge of Experiences experiences Knowledge of trade off costs with effective (wilderness value (economic, social, decision making experience) positions etc.) and action t FOR A.M E O F R E F E R E N C E ‘9 An evolving frame of reference (pre-existing values, beliefs and attitude systems) accommodatesanuiassimilates new knowledge and experiences (Beliefs concerning (Beliefs concerning (Beliefs concerning (Beliefs concerning desirability of consequences of the tactics of env. own ability to land preservation) land preservation) action taking) bring about change) \\\\\\\\\M ¢//////// A_T T I T U D E 3 \\\\\\\\\\\¢/////////// A1 A2 (Attitude concerning the (Perception of ability to importance - value - of bring about change in this preserving Alaskan lands) type of situation - LOCUS OF CONTROL) \\\\g If Al is weak If A1 is strong If A2 is INTERNAL If A2 is EXTERNAL AND THEN 1» l’ w Low probability High probability of Low probability of action action (e.g., letter of action writing to senator re Alaskan land bill) G O A L O F E E Figure 2.3. 39 experiences with, environmental problem solving as reflected by the impinging variables on the upper right side of the model... in addition, L of C implies that other beliefs (B3 and BA) and attitudes (A2) are important considerations in bringing about a specific environmental action. The processes involved as precursors to an environ- mental action are implied by the frame of reference component. The paradigm assumes that a citizen has a frame Of reference which reflects all past learning experiences, values, beliefs, and attitudes, and which serves to process any new knowledge and/or experience. Some of this new in- put is modified to "fit" into the existing frame of reference (assimilation). In other cases, the frame of reference itself is adjusted to accept new perspectives (accomodation). The result is a constantly evolving frame of reference com- prised of new beliefs and attitude systems. The attitude systems prevailing at any given time will determine the types of behaviors, if any, that are engaged in (pp. 183-185). With the exception of some episodic studies based on the use of generalized L of C scales little effort has been expended by in- vestigators to validate the proposed relationship between environmental action taking and L of C beliefs. Peyton and Miller (1980) recommend the development of "valid and reliable instruments which will measure situation specific (i.e., L of C in environmental problem solving) rather than generalized expectancy of reinforcement" (1980, pp. 185). As previously discussed, Levenson (1972b) found that males who believe that chance did not control their lives were involved in signi- ficantly more anti-pollution activities than were those who perceived that chance had more control of their lives. Similarly, male members of an anti—pollution group scored significantly lower on the C scale (more internal) than did non-members. In addition, those who believed in chance were less knowledgeable about pollution than their less 40 chance—controlled counterparts. Although Levenson's instrument assesses three separate belief systems (I, P, C), it samples across a variety of life's experiences and is therefore identified as a generalized L of C instrument. As the literature has shown, a generalized instrument may result in weak findings when applied to a specific situation. It is possible Levenson could have improved on her findings if the I, P, C items addressed the specific situation rather than life's ex- periences in general. Arbuthnot (1977) developed a Study to examine the social and personality profiles of recyclers (N=85 known users of a recycling center) 4 and non—recyclers (N=6O conservative rural church members). The questionnaire administered to the group measured demographics, environmental behavior, attitude/personality and environmental know— ledge. A number of questions in the personality/attitude section assessed personal control or L of C beliefs. It was found that the four best predictors of the use of a re- cycling center were education, environmental knowledge, general con— servatism and lack of personal control (i.e., L of C). Thus, a recycler in this study could be described as a person relatively well educated, knowledgeable about environmental issues, relatively liberal in political, social and religious beliefs and who feels he has some potential impact by his actions. Recyclers were more likely to score lower on Lack of Personal Control questions indicating their perceived ability to exercise control over events in their own lives. Recyclers were more concerned about the future consequences of present policies and felt more compeled to take action. At the same 41 time recyclers felt that their individual actions may have little long-range impact in the face of the ecologically unsound activities of large corporations and government agencies. Caution, however, must be exercised in reviewing Arbuthnot's results. Personal Control questions (L of C questions) were extracted from a variety of different L of C instruments which could have an effect on reliability and vali- dity of the findings. In a similar study, Tucker (1978) found that a predictive model consisting of internal—external control (as measured by the I-E scale) social responsibility, age, income, and social class was capable of significantly discriminating between members of a Sierra Club and/or Audubon Society group and the general population. Based on results Tucker identified Sierra Club and/or Audubon Society membership as a relatively "effective operational definition of environmental respon— sibility" (pp. 410). The model also proved to be an effective dis- criminator of high and low environmental responsibility in the general population. More specifically, it was found that individuals who under- take environmentally oriented activities perceive themselves as being in control of their life experiences. Although findings indicate an initial linkage between L of C and environmental responsibility, Tucker recommended that "further research efforts should be concerned with modifying the internal—external control scale to represent more environmentally specific items" (pp. 415). Smith (1980) was concerned with validating a newly—developed environmental L of C instrument (Tomera 1979) and with exploring the 42 possible relationships that might exist between a measure of environ— mental participation and L of C beliefs. The four groups of college students used in the study included in-service elementary teachers N=21), an environmental interpretation class (N=22), an introductory science methods class (N=17) and a grOup of environmental education students (N=43). Smith found that the environmental education groupperceived them- selves to be significantly more involved in environmental actions then did the other three groups. No significant differences were found among the elementary teachers, environmental interpretation class, or the science methods class. Smith pOints out that these findings are supported by Bluhm (1979). He found environmental educators to be: more involved in environmental action than environmental coordinators, pre—service teachers and the general public. Further data analysis from the Smith study indicated a significant correlation between scores on the Rotter I-E scale and the Environment L of C scale for the elementary teachers group, the interpretation class and for the science methods class. Based on these findings Smith inferred that both instruments measured the same construct. However, she questioned whether either instrument was measuring L of C with re— spect to environmental action. Significant low level negative correlations were found between I-E scores, Environmental L of C scores and perceived environmental action taking, only for the environmental education group. Thus, the Environmental L of C scale, which was intended to be more sensitive than the I-E scale, provided the same results. This would seem to indicate that little predictive power was gained by developing the Environmental L of C scale. 43 The Environmental L of C scale was a brief, four item, forced choice instrument. Validity questions arise since item and/or whole test reliability was not determined. Due to the rather inconclusive findings, Smith (1980) recommended further research be conducted with a more comprehensive valid and reliable situation specific instrument. Miller (1981) conducted a study to determine if a Youth Conservation Corps experience had an effect on the participants' L of C. Levenson's I, P, C scales were used to measure the participants‘ L of C before and after the experience. Findings indicated no evidence that the YCC experience in general influenced L of C development in participants. T However, positive correlations were found between reported number of environmental actions taken and internality. Based on these results Miller recommended the development of a L of C measure specific to environmental action situations so that the indicated relationship between L of C and environmental action taking behavior might be better investigated. Literature Review Summary Locus of Control constitutes a personality dimension that can be quantified and used along with other variables to explain and/or predict human behavior. The concept of L of C was originally intro— duced as one of four equally weighted variables in Rotter's Social Learning Theory and it can best be understood within the context of this theory. The literature proposes that L of C may be one of many variables which impinge on the environmental action taking behavior of an individual. 44 Therefore, L of C research findings may have implications for the design of more effective environmental education programs. Those research studies that indicate a relationship between L of C and environmental action taking have been brief and/or based on the use of generalized instruments which are subject to predictive short- comings when applied to specific situations. Researchers have recommended that a L of C instrument specific to environmental action be designed and used to further investigate the prOposed relationship between L of C beliefs and environmental action taking behavior. Developing and refining situation specific instruments which discriminate between different external beliefs (e.g., powerful others vs chance) have proven fruitful. Incorporating these same refinements to further investigate L of C's relationship to environmental action taking behavior may also prove worthwhile. Chapter 3 Research Procedures The purpose of this study was to construct a Locus of Control (L of C) instrument which could be used to measure the perceived expec— tancy of an individual for taking specific types of environmental action in a stated situation. In addition, this instrument was tested for initial evidence of reliability and validity. The instrument de— signed in this study is entitled the Perceived Environmental Control Measure (PECM). This chapter describes procedures used in instrument design, pilot and field testing of the instrument, as well as methods used for data analysis. Instrument Design The final PECM is composed of two major sections. The first section contains the summary of an environmental issue and the second section contains the PECM statements. In this study a third section was attached to the PECM. This section contained questions on the ex— tent and types of environmental actions actually taken by the respondent. This data was necessary to assess contruct validity of the PECM. Further rationale for the content development and inclusion of the various sections is presented below. 45 46 Section I — The Environmental Issue In section I of the.PECM, respondents are asked to read a summary of an environmental issue and to consider themselves as directly in— volved in the issue. After reading the issue, the subject is directed to respond to the PECM statements. Rationale for inclusion of the issue is based on Rotter's Social Learning Theory (SLT). According to SLT the probability of a given behavior is a function Of the expectancy that a reinforcement will follow, the value of the reinforcement, and the nature of the specific situation in which the behavior is to occur. The environmental issue summary places the reader in a "specific Situation." The reader is also confronted with a "reinforcement" (i.e., environmental quality outcome of the issue) that Should be of some concern or value to the individual. In addition, the issue is designed to make it possible for an individual to apply any or all five categories of environmental action which are synonymous with the behavior component of Rotter's SLT. Thus, three classes of variables from the SLT are addressed by the environmental issue summary: 1) the specific situation; 2) the valued reinforcement; 3) a potential class of behavior. Section II — PECM Statements Section II of the PECM is designed to measure the final component of Rotter's SLT, which is the "expectancy" for reinforcement (i.e., desired outcome) through use of the various environmental actions (i.e., behaviors). — 47 It could not be assumed that the respondent would identify or associate the use of all environmental action categories with the situ- ation described in Section I. To be assured that the respondent was aware of the actions that could be taken a definition of each is pro- vided in Section II of the PECM. Following each environmental action definition are a series of L of C statements that pertain to the use of that action in the given situation. Item Content and Scale Construction of the PECM ) Initially, an item pool was developed with 150 statements divided equally among three L of C belief orientations - Internal (I), Powerful Others (P), and Chance (C). These belief orientations were applied equally across five categories of environmental action — political action, persuasive action, economic action, and ecomanagement (environmental action categories are defined in Chapter 2 and Appendix B). Examples of I, P, and C statements as they related to one of the five environmental action categories are presented in Figure 4. Letters in the parentheses indicate whether the item represents I, P, or C orientation. Developing the PECM scale using the I, P, or C orientations and the five categories of environmental action as content parameters allows the instrument to capitalize on two refinements in I—E measurement that have proven fruitful. First, identifying and directing ex— pectancy statements toward environmental actions allows the instrument to be situation specific. That is, it addresses behaviors that are 48 POLITICAL ACTION: An eééont aimed at pehéuading an ezeetonare, a. Keg/Osman (on Zegutatwte), OIL executive goue/znmen/tat agency to confiolun to the wine/s heed by the roe/won on pen/50m taking that action, e.g., tabbying, vottng, cam- paigning 601 candidates, etc. (C) I believe that what is going to happen in this situation will happen regardless of any political action I take. (I) By participating in some type of political action, I can play an effective role in determining the outcome of this situation. (P) The political action I could take in this situation would be of little value in determining the outcome, because the outcome will mostly be influenced by a few peOple who already have their own ideas about the situation. Figure 3.1 Examples of Internal (I), Powerful Others (P) and Chance (C) statements as they relate to Political Action.\ 49 specific to the situation described. Second, using the tripartite division of I, P, C allows the instrument to be more sensitive to the different external orientations that were identified by Levenson (1972b). PECM Scale Construction Scales are devices constructed or employed by researchers to quantify the responses of a subject on a particular variable (Tuckman 1978). The scale employed in quantifying responses to PECM items is a six-point Likert-type scale. A number of researchers support the use of a Likert-type format in measuring attitudes or beliefs. An important study concerned with the effectiveness of different attitude scales was conducted by Tittle and Hill (1967 in Borg and Call 1979). It was reported that this study compared the effectiveness of various scales (i.e., Likert, Guttman, Sematic Differential, Thurstone, Self Rating) in predicting objective indices of voting behavior. The Likert scale was found to be superior to all other scales tested; it yielded a mean correlation coefficient of .54 with objective indices of voting behavior. Oppenheim (1966) also supports the use of a Likert-type scale. He states that the Likert scale tends to yield "good" reliability because of the wide range of answers permitted to respondents. Millward (1975, pp. 50 in Burrus-Bammel, 1978, pp. 44) states that Likert-type scales are ”easy to construct and administer" as well as being "valid and reliable in measuring attitudes towards a variety of environmental topics." 50 Scoring the PECM In the PECM, the P and C items are written in the external direction and items in the I scale are written in the internal direction. The P and C items were scores by the following key: Disagree strongly = Disagree somewhat Disagree slightly Agree slightly Agree somewhat Agree strongly = II O‘U‘IDUJNl—I The I items were scored by the following key: Disagree strongly = Disagree somewhat = Disagree slightly = Agree slightly = Agree somewhat = Agree strongly = h—‘NQDUIO‘ The I item scoring is reversed from P and C item scoring. This reversal is necessary so that all three scales reflect a common direction with regards to the externality and internality continuum. A relatively high score on the I items will reflect a low belief in that orientation. A relatively high score on the P or C items will reflect a high belief in those orientations. A relatively low score on the I items will reflect a high belief in that L of C orientation; whereas a relatively low score on P and C items reflects a lack of belief in those orientations. Analysis of data from Likert—type scales are usually based on summated scores. To provide a manageable yet reliable and valid in- strument it was proposed the final PECM contain 45-I, P, C items applied equally across the five categories of environmental action. Three types 51 of summated scoring systems were applied to these 45 items and tested for utility. These scoring systems included individual I, P, and C scores summed across all environmental action categories; combined I+P+C scores for-each environmental action.category; and, combined I+P+C scores summed across all environmental action categories (total PECM score). Table 3.1 presents the number of items and range of scores for the scoring system that recognized individual I, P, and C scores summed across all environmental action categories (total I, P, or C score). Based on this scoring system an individual received three scores - an I, a P and a C score. The range of each score falls on or between nine - (internal orientation) and 45 (external orientation). This system of scoring differentiates between the external orientations of powerful others and chance and is similar to the tripartite analysis proposed by Levenson (1972b). The second scoring system considered and tested for utility was the combined I+P+C scores (total action category scores) that were achieved by an individual for each action category. Based on this system the individual received five scores. The number of items in— volved in each score and the range of each scores is presented in Table 3.2. This type of scoring system falls back on the unidimensional approach to L of C measurement. That is, a person may exhibit a common tendency to score internal or external about the personal use of environ- mental action in a given Situation. Differentiation between different types of external orientations is not considered in this scoring system. A relatively high combined I+P+C score could be assumed to reflect an external orientation toward the particular action category in which it was achieved. 52 Table 3.1. Number of Items and Range of Scores for Total I, P, and C Variables in the PECM. Variable Range of Scores Internal (I) Score Chance (C) Score Powerful Others (P) Score 15=3 I's (5 action categories) 15=3 C's (5 action categories) 15=3 P's (5 action categories) 45 Total Items Internal External 15 —————————— 90 15 ---------- 90 15 —————————— 90 1 53 Table 3.2. Number of Items and Range of Scores for Total Action Category Variables in the PECM. Range of Scores Variable Total Number of Items Internal External Political Action 9 (3 I + 3 P + 3 C) 9 ___________ 54 Score Persuasive Action 9 (3 I + 3 P + 3 c) 9 ___________ 54 Score Ecomanagement 9 (3 I + 3 P + 3 c) 9 ___________ 54 Score Economic Action 9 (3 I + 3 P + 3 c) 9 ___________ 54 Score Legal Action 9 (3 I + 3 P + 3) 9—____-__--_54 Score 45 Total Items 54 A third type of scoring system being considered and tested for utility is the combined I+P+C score that can be achieved by an individual across all action categories (i.e., total score). The number of items involved in this scoring system and the possible range of scores is presented in Table 3.3. Based on this system the respondent will re— ceive only one score for the entire PECM. This scoring system also reflects the unidimensional approach to L of C measurement. A relatively low total score might be assumed to reflect internality toward taking environmental actions in general. A relatively high total score might be assumed to reflect externality toward taking environmental actions 1 in general. Section III — Citizen Action Questions The third section attached to the PECM consists of questions de— signed to assessan individual's actual - perceived use of political, persuasive, economic, ecomanagement and legal actions (Appendix D). This section is not intended to be a permanent part of the final PECM. It is included in this study to test for evidence of instrument validity. Theoretically those who score internal on the PECM should also score relatively high on the environmental action questions (this hypothesis is discussed further in the validity section of this Chapter). Scoring Citizen Action Questions One summated score was used to quantify the action taking reported by each respondent. An individuals final summated score for all citizen action questions could range from 0—121 points. A series of citizen action questions required a numerical response from the subject. 55 Table 3.3. Number of Items and Range of Scores for Total PECM Scores. Range of Scores Variable Total Number of Items Internal External Total Score 45 = 15 I + 15 P + 15 C 45 —————————— 270 45 Total Items 56 That is, the subject was asked to state the number of times he/she had taken actions within a certain time period. Each action cited was worth one point. The maximum on one question was Six points. An example (Part A) is presented below. How many times during the past two (2) years have you... A) ...contacted a politician to express your support or opposition to a bill they have introduced or are considering for passage. 3 (no. of times) 1 B v (Please list up to three (3) titles or topics of such bills below). 1. Bottte W 2. Manhan Lands W 3. In this case, the individual responded three so he/she received three points. A response of eight would have received the maximum of six points. If no response was provided zero points would be received. As a further check on activism subjects were asked to list up to three examples of the actions they claimed to have taken (Part B above example). Subjects received two points for each example of an action they listed. In Part B of the above example, the individual listed two actions and received four points. If no activities had been listed, zero points would have been received. 57 In the final two citizen action questions (Section III), the respondent was asked to only check (y’) responses that he/she per- ceived to describe actions he/she had taken. These questions are re- stated below. Have you used your right to vote in an attempt to improve situations (issues) about which you are concerned? *’ Yes No Please place a check mark (V’) in front of each activity you have participated in over the last two (2) years. ‘ l. I have picked up litter and/or organized a litter campaign. 2. I have taken steps to reduce energy con- sumption. 3. I have avoided the purchase of a product because of its negative effect on the environment. 4. I have taken steps to reduce my water consumption. 5. I have recycled paper, glass, metals and/ or organic refuse. 6. I have participated in a habitat improvement project (e.g., planting shrubs for wild- life, putting up birdhouses, stream re- novation). In the first question the individual received one point for checking "yes" to using the "right to votefl recieved zero. worth one point. 0-6. ' A "no" response would have In the second question each action checked was also The range of point values for this question was 58 Validity and Reliability of the PECM Establishing Evidence of Content Validity Rather than test the validity of measures after they are constructed, one should ensure validity by the plan and procedures of construction (Nunnally 1978, pp. 72). Two approaches were pursued to establish initial evidence of content validity for the PECM. First, an item pool was developed based on a specific set of objectives. Second, items were selected from the item pool, for the final instrument based on relatively high item subscale and/or item total correlations. Borg and Call state that "content validity is determined by syste- matically conducting a set of operations such as defining in precise terms the specific content universe to be sampled, specifying objectives, and describing how the content universe will be sampled to develop ! items.’ The content universe of PECM items is based on I, P, C Locus of Control orientations applied to five categories of environmental action (i.e., political, persuasive, ecomanagement, economic, and legal action). Specific objectives applied to development of the PECM items are presented below. 1. Items in the I-scale will be constructed to elicit responses which measure the degree to which an individual perceives his/her use of an environ- mental action will have an effect, or control, the outcome of a stated situation. 2. Items in the P—scale will be constructed to elicit responses which measure the degree to which an individual perceives powerful others, more than his/her own use of an environ- mental action, will control or have an effect on the outcome of a stated situation. 59 3. Items in the C-scale will be constructed to elicit responses which measure the degree to which an individual perceives chance or fate, more than his/her use of an environmental action, will control or have an effect on the outcome of a stated situation. Based on these objectives an item pool of 150 statements was de- veloped and through pilot studies item total correlations (how well an item correlated with all other items that are proposed to measure the same variable) were calculated. Items with relatively high item total correlations were retained for the final PECM. This method of item selection is recommended by Nunnally (1978), Tuckman (1978) and Oppenheim (1966); it should yield a measure with a set of rather homo— geneous items (i.e., items that measure something in common) which supports evidence of content validity. Establishing Evidence of Copptruct Validity Construct validity is defined as "the extent to which a particular test can be shown to measure a hypothetical construct" (Borg and Call 1978, pp. 216). To obtain information needed to establish evidence of construct validity, the investigator formulates hypotheses about those who have high scores on the instTument in contrast to those who have low scores. These hypotheses or theoretical formulations lead to certain predictions about how subjects at different score levels on the instrument will behave on certain other tests or in certain defined situations. If the investigators' theory about what the in— strument (developed from the construct) measures is correct most pre— dictions should be confirmed and evidence of construct validity is supported (American Pyschological Association 1974). 60 As discussed in Chapter 2, researchers have shown some indication of a relationship between internality and environmental action taking behavior. Theoretically, similar findings should be attained with the PECM. That is, the more action taking an individual is involved in, the more he Should score in an internal direction on the PECM. To investigate this hypothesis, scores on a series of questions regarding actual-perceived use of political, persuasive, economic, legal and ecomanagement type actions were correlated with PECM scores. To support evidence of construct validity those who score internal on the PECM should score relatively high on the question regarding perceived use a, of the various actions. Establishing Reliability of the PECM Coefficients of internal consistency will be calculated for all scoring systems being considered. The method used for estimating in- ternal consistency willbe Cronback's Alpha which yields an estimate of reliability based on average correlation among items within the in- strument. Nunnally (1978) states that this type of reliability test should be applied to all new measurement methods. Pilot Studies I and II Initially, a PECM item pool was developed and two pilot studies were run with the primary objective of analyzing and selecting valid and reliable items for a final instrument. A description of each pilot study is presented below. 61 Pilot Study I The primary objective of pilot study I was to select a relatively internally consistent group of items for pilot study II. A total item pool of 150 statements was developed for the first pilot study. These items were divided into five separate instruments based on the five categories of environmental action. Each of the instruments contained the same environmental issues summary and 30 I, P, C statements directed toward the use of a particular environmental action in the given situation (issue). The I, P, C statements were randomly arranged (using a random number table) in each of the instruments. Figure 5 presents a listing of the five instruments. Each instrument listed in Figure 32 was administered to a different sample of undergraudate students (N>50). These samples were drawn from three different universities and were selected based on avail— ability. Each sample contained a diverse background of college majors. Results of this study are presented in Chapter 4. Pilot Study II Three objectives were set for pilot study II: 1) analyze and select reliable items for the final PECM; 2) investigate the reliability of the various scoring systems being considered; 3) test the PECM for evidence of construct validity. With the exception of being longer, the instrument developed for pilot study II was intended to have the same format as the final PECM. An outline of the instrument is presented below (the complete in— strument is provided in Appendix B). 62 Instrument Variable 1. Ecomanagement 2. Persuasive Action 3. Political Action 4. Economic Action 5. Legal Action Total Items 30 IPC statements 10 I + 10 P + 10 30 IPC statements 10 I + 10 P + 10 30 IPC statements 10 11+ 10 P + 10 30 IPC statements 10 I + 10 P + 10 3O IPC statements 10 I + 10 P +>10 O O O Figure 3.2. Listing of the Five Instruments Used in Pilot Study I. 63 I. Introduction II. Section I A. Summary of an environmental issue III. Section II A. Definition of each environmental action category B. Each definition precedes 15 IPC = SI + 5P + 5C statements directed toward the particular en- vironmental action defined C. Total number of IPC statements in the in— strument is 90 = 5 action categories x 15 statements per/category. IV. Section III A. Environmental action questions 1. Total of 21 questions designed to assess actual perceived amount of environmental actions taken. Instruments in the above format were administered to two classes of undergraduate students at Michigan State University (MSU). The first class consisted of students enrolled in a course designed to survey environmental issues (N=91). The second class consisted of students enrolled in an introductory course in fisheries and wildlife management (N=29). Results of this study are presented in Chapter 4. Field Testing the PECM The PECM developed from the two pilot Studies was subjected to four field tests. The intent of these field tests was to assess in- strument reliability and validity and to explore the relationships that might exist between perceived environmental action taking and L of C. This section presents descriptions of the final PECM(s), 64 the four groups it was administered to, and the research questions which were posed. Description of PECM The PECM(s) used in the field tests were of a format similar to the longer instrument used in pilot study II. However, the final in— strument(s) were only 45 IPC items rather than 75 and included demo— graphic questions. Also, in one of the field tests, the effects of two different environmental issue summaries was investigated. The solid waste issue used in pilots I and II was given to half of a 1 selected group, while the other half of the group received an acid rain issue summary (Appendix C). Sample Sources and Procedures for Data Collection Field testing of the PECM involved collecting data from four groups of participants: 1) undergraduate students in an interdisciplinary environmental issues survey course; 2) undergraduate students in an introductory biology course for non-science majors; 3) members of the Central Michigan Sierra Club (MSU); and 4) a sample Of K—5 teachers selected from 20 school districts in the Grand Rapids, Michigan area. Group I: This group consisted of undergraduate students enrolled in a course entitled Resource Ecology and Man (N=85). The course presents a survey of environmental issues and may be taken as an elective to meet part of the basic natural science requirements of several programs at MSU. Students in the course represent freshman, sophomore, junior and senior levels as well as a variety of majors. 65 This class was chosen based on its large enrollment, students' diverse backgrounds and availability. The PECM was administered to the group during their first week of classes (spring term). Students were given in-class time to fill out and return the instrument. Group II: This group consisted of undergraduate students enrolled in an introductory biology course for non-science majors (N=78). Approximately two thirds of the class consists of education majors and the final third represents a variety of non—science majors. The group was chosen based on large enrollment and availability. This group was divided into two subgroups. Half of the group (group II A) was given a PECM with the solid waste issue summary. The other half of the group (group II B) was given a PECM with the acid rain issue summary. Students were given in class time to fill out and return the instruments. To assure some randomness in dis- tribution of the two issues they were handed out on an every-other- seat basis. ' Group III: This group consisted of approximately 23 members of the Michigan State University/Central Michigan, Sierra Club who volunteered to participate. This membership is Often involved in taking actions on various environmental issues. Individual members come from a variety of backgrounds. The PECM was distributed at the Club's monthly meeting. Members were asked to fill out and return the instrument within seven days. Group IV: This group consisted of a sample of K-S teachers (N=159) selected from 20 school districts in the western Michigan area. The group was divided into two subgroups. Fifty-nine 66 of the teachers (group IV A) have worked with an area environmental education consultant and have implemented environmental education pro- grams in their classrooms. The other 100 teachers (group IV B) repre— sented the faculties of 10 schools selected randomly from a list of schools in the same area that have not worked with the environmental education consultants. The instrument was distributed by mail. Teachers were asked to fill out and return the instrument within seven days after receiving it. Research Questions i The purpose of this study is to answer the following research questions. A series of the questions are followed by research hypotheses (i.e., questions 2, 3, 4, 5). These hypotheses are posed to test the PECM for evidence of construct validity. The remaining research questions are exploratory in nature. ' Research Question 1. Can evidence of reliability and validity (content and construct) be established for an instrument developed to measure perceived L of C toward taking specific categories of environmental action in a stated situation? Research Question 2. What relationship exists between scores on a measure of citizen action and total PECM scores within the environ- mental issues class, the Sierra Club group, each of the teacher subgroups and each of the introductory biology class subgroups? Research Hyppthesis 2.1. A significant (P.i .05) correlation will exist between scores on a measure of citizen action and total PECM within each of the selected groups. 67 Research Question 3. What relationship exists between scores on a measure of citizen action taking and total I, P, and C scores for the environmental issues class, the Sierra Club group, each of the teachers subgroups and each introductory biology class subgroup? Research Hyppthesis 3.1. A significant negative correlation (p §_.05) will exist between scores on a.measure of citizen action and total I scores for each selected group. Research Hypothesis 3.2. A significant negative correlation (pq: .05) will exist between scores on a measure of Citizen action and total P scores for each selected group. Research Hypothesis 3.3. A Significant negative correlation (p_: .05) will exist between a measure of citizen action and total C scores for each selected group. Research Qpestion 4. What relationship exists between scores on a measure of citizen action and total I+P+C scores for each environ- mental action category within the environmental issues class, the Sierra Club group, each.of the teachers subgroups and each of the introductory biology subgroups? Research Hypothesis 4.1. A significant negative (p :_.05) correlation will exist between scores on a measure of citizen action and total I+P+C scores for each environmental action category within each of the selected groups. Research Question 5. What relationship exists between total I, P, and C scores within the environmental issues class, the Sierra Club group, each of the teachers subgroups and each of the introductory biology subgroups? g 68 Research Hypothesis 5.1. A significant positive (p §_.05) correlation will exist between total I, P, and C scores within each of the groups selected (a relatively high I score refers to lack of belief in internal control.) Research Question 6. Is there a significant difference between total I, P, or C scores within the environmental issues group, the Sierra Club group, each of the teacher subgroups and each introductory biology subgroup? Research Hypothesis 6.1. A significant difference in mean I, P, and C scores will be determined within sample groups. Research Question 7. IS there a significant difference(p §_.05) by sex, between environmental action category scores within the en- vironmental issues group, the Sierra Club group, each teacher subgroup and each of the introductory biology subgroups? Research Qpestion 8. DO biology subgroups differ significantly (p_: .05) on total PECM scores, I, P, C scores, total environmental action category scores and/or Citizen action scores? Research Question 9. Do teacher subgroups differ significantly (p_: .05) on total PECM scores, I, P, C scores, total environmental action category scores and/or citizen action scores? Research Question 10. Does the environmental issues class and biology subgroup II A differ significantly (P.i .05) with respect to total PECM scores and/or citizen action scores? 69 Research Question 11. Based on total PECM scores and/or citizen action scores, do teachers as a group (groups IV A and B) differ significantly (p :_.05) from college students administered the solid waste version of the PECM? Research Question 12. Do total PECM scores and/or citizen action scores for the Sierra Club differ significantly (p :_.05) from a combined sample of all other subjects who responded to the solid waste version of the PECM (i.e., groups I, II A, IV A, IV B)? Data Analysis The Office of Research Consultation at Michigan State University (aided in developing the analytical procedure, data preparation and computer programming. Analytical procedures were accomplished using the Statistical Packages for the Social Sciences (SPSS). Various parametric statistics were utilized to examine the re- search questions posed in the previous section. The Likert-type scale, as employed in the PECM, is characterized as an interval scale (Tuckman 1978). Borg and Gall (1979) state that when data meet the assumption of being interval scores, it is advisable to use parametric statistics. When using interval scores, moderate departure from theoretical assumptions has been shown to have very little effect upon the value of parametric statistics. Cronback's alpha reliability coefficients were calculated for the total PECM and for each of the various subscales. In addition, these coefficients were calculated independently for data collected 70 from each teacher subgroup, the environmental issues class, the Sierra Club group and for each biology class subgroup. Research questions 2—6 along with corresponding hypotheses are based on L of C/IPC theory and were tested to determine if the PECM exhibits evidence of construct validity. Research questions 7-12 were designed to further explore relationships that might exist between environmental action taking behavior and L of C beliefs. Specifically, research questions 2-5 addressed an investigation of relationships between certain variables. A Pearson's product—moment correlation (Pearson's r) was employed for analysis. The Pearson's r can be used to determine the degree of relationship between two sets of scores (i.e., variables). Pearson's r is applicable to analyzing interval scores (Tuckman 1978) and/or continuous scores (i.e, when scores on one variable can theoretically occur at any point along a continuunD(Borg and Gall 1979). Both of these characteristics apply to the Likert—type scale used in the PECM. Significance of the various coefficients were determined at the .05 level. Research questions 6—12 addressed an investigation of significant differences between certain scores. A series of analysis of variance programs was employed to explore these research questions. Specifically, if differences between environmental action category scores or IPC scores, by sex, or by group, were being examined an SPSS MANOVA Profile Analysis (repeated measures) program was employed. This program addresses three questions. First, are the population mean profiles similar, in the sense that line segments of adjacent tests are parallel? 71 Second, assuming parallelism, are the profiles of sample groups equal. Third, assuming parallelisun are the response means for the various tests equal. If differences between total PECM scores or citizen action scores were being examined, a one-way analysis of variance was computed using SPSS MANOVA and Planned Contrasts when appropriate. Chapter 4 RESULTS OF DATA ANALYSIS The results of data analysis for the two pilot studies which were used to develOp the final Perceived Environmental Control Measure (PECM) are presented in this chapter. In addition, the final PECM was sub- jected to a series of field tests. The intent of these field tests was to assess the instruments' reliability and validity and to further explore relationships that might exist between environmental action taking behavior and Locus of Control beliefs. Pilot Studies I and II Data Analysis Results The primary objective of pilot study I was to select a relatively homogeneous group of items for pilot study II. Initially, an item pool of 150 PECM statements was developed and divided equally into five separate instruments (Figure 3.2). Each instrument was administered to a separate sample of college students (N >50). Criteria used for the retention of items for pilot study II were relatively high item totoal correlations (i.e., how well the item correlated with the other 29 items of the respective instrument) and relatively high item subscale correlations (i.e., how well the item correlated with the other items of the same L of C orientation, i.e., how well a C item correlated with the other nine C items of 72 IIIIZ;___________________________________________________________________________lL_--IIIIIII 73 the instrument). As discussed in Chapter 3, this method of item selection is recommended by Nunnally (1978), Tuckman (1978) and Oppenheim (1966). Based on the preceding analysis, 15 items (SI, 5P, 5C) were se- lected from each of the five instruments. All items selected had a Pearson product moment correlation coefficient (Pearson's r) of greater than .50 with their respective total and subscale scores. In all, a total of 90 items were selected from pilot study I and collapsed into one instrument for pilot study II. The second pilot study differed from the first in that a subject responded to all five categories of environmental action. This second pilot study also included a series of questions designed to assess an individual's previous use of political, persuasive, economic, ecomanagement, and legal actions (citizen action questions). These questions were included to test the PECM for evidence of construct validity (see discussion on validity Chapter 3). Instruments were.administered to two classes of undergraduate students at Michigan State University (MSU). The first class consisted of students enrolled in a course designed to survey environmental issues (N=91). The second class consisted of students enrolled in an introductory course to fisheries and wildlife management (N=29). The first objective of pilot study II was to select a homogeneous set of items for the final PECM. Item selection was again based on item total and item subscale correlations. However, in this study "item subscale" refers to the I, P, and C scores totaled across all environmental action categories. In pilot study I "item subscale" 74 referred to the I, P, and C scores within a given environmental action category. A total of 45 items (15I + 15P + 15C) were selected for the final PECM. All the 45 items had a Pearson's r of greater than .55 with their respective total and Subscale scores. A second objective of pilot study II was to determine reliability coefficients for the various scoring systems being considered. Cronback's alpha reliability coefficients (a measure of internal consistency) were calculated for all but one of the PECM scoring systems being considered. The reliability coefficient for the total score (i.e., combined I+P+C score summed across all environmental action categories) was not calculated 1 a due to an oversight in programming. However, this coefficient was determined in the final field testing of the PECM. Nunnally (1978, pp. 245) states that ”in the early stages of research on predictor tests or hypothesized measures of a construct... reliabilities of .70 or higher will suffice." All of the reported reliabilities for pilot study II were above .84 (Table 4.1) which indicates that the various scoring systems being considered Show evidence of high internal consistency (reliability). A third objective Of pilot study II was to test the various PECM scoring systems for evidence of construct validity. It was hypothesized that subjects who scored relatively internal on the PECM would also score high on citizen action questions. To investigate the relationship between the various PECM scores and the citizen action question scores, a series of Pearson product—moment correlation coefficients (Pearson's r) were calculated. The Pearson's r can be used to determine the degree of relationship between two sets 75 Table 4.1 Pilot Study II Reliability Coefficients for PECM Scoring Variables. Variables Alpha Total I, P, and C Scores Chance (C) .95 Internal (I) .96 Powerful Others (P) .93 Total Action Category Scores Ecomanagement .84 Economic .95 Political .93 Persuasive .95 Legal .96 76 of scores (i.e., variables). Pearson's r is applicable to analyzing interval scores (Tuckman, 1978) and/or continuous scores (i.e., when scores on one variable can theoretically occur at any point along a continuum; Borg and Call, 1979). Both of these characteristics apply to the Likert type scale used in the PECM. Significance of the various coefficients were determined at or below the .05 level. For each of the various PECM scoring systems being considered, a significant negative correlation (P.i .05) was achieved with citizen action question scores (Tables 4.2, 4.3, 4.4). Individuals with relatively internal scores (low IPC scores) perceived themselves to be taking more actions (high environmental action taking scores) than their comparatively external counterparts. This finding supports the proposed hypothesis and indicates evidence of construct validity for each of the PECM scoring systems used in pilot study II. Field Testing Data Analysis Results Field testing of the PECM involved collecting data from four major groups of participants. Group I consisted of undergraduates enrolled in an introductory environmental issues course (N=85). Students were given time in class to fill out and return the instrument. A total of 81 =usable instruments were returned. Group II consisted of undergraduate students enrolled in an intro- ductory biology course for non-science majors. This group was divided into two subgroups. A total of 40 students (subgroup II A) were given the PECM with the solid waste issue summary and a total of 41 students (subgroup II B) were given a PECM with the acid rain issue summary. The appropriate instruments were handed out during lab 77 Table 4.2. Pilot StudyII, Pearson Product—Moment Correlation Co— efficients for Total I, P, and C with Citizen Action Scores. Correlation Significance Sample Variables Coefficient Level Size Correlated .5 .p .N Total Internal (I) Scores/Citizen Action Scores -.4342 .001* 116 Total Chance (C) Scores/Citizen Action Scores -.4076 .001* 117 Total Powerful Others (P) Scores/ Citizen Action Scores -.3418 .001* 117 * Significant Relationship (p §_.005). 78 Table 4.3 Pilot Study II, Pearson Product-Moment Correlation Co- efficients for Total Action Category Scores and Citizen Action Scores. Variables Correlated Correlation Significance Sample with Citizen Coefficient Level Size Action Scores .p p .5 Total Political Action Scores —.3816 .001* 120 Total Persuasive Action Scores —.4571 .001* 118 I Total Eco- management Scores -.2131 .010* 119 Total Economic Action Scores -.3139 .001* 117 Total Legal Action Scores -.3454 .001* 117 * Significant relationship (p §_.05). 79 Table 4.4 Pilot Study II, Pearson Product-Movement Correlation Co— efficient for Total PECM Scores and Citizen Action Scores. Correlation Significance Sample Variables Coefficient Level Size Correlated r p N Total PECM Scores/Citizen Action Scores —.3917 .001* 120 * Significant relationship (P_: .05). i, 80 periods on an every—other—seat basis to provide some randomness in distribution. Students were given time in class to fill out the in- struments. A total of 38 usable insturments were returned from sub— group II A and a total of 40 usable instruments were returned from subgroup II B. Group III consisted of members from the Michigan State University/ Central Michigan Sierra Club. A total of 23 PECMs were distributed at a monthly meeting. Members were asked to fill out the instrument at home and return it by mail. A total of 14 instruments were returned. Ten of these instruments were usable. Four of the instruments were discarded i because reSpondents were not Sierra Club members. Group IV consisted of a sample of 159 K-5 teachers selected from 20 school districts in the western Michigan area. This group was divided into two subgroups. Fifty—nine of the teachers (subgroup IV A) have worked with an intermediate school district environmental education coordinator and are known to have implemented environmental education programs in their classrooms. The other 100 teachers (subgroup IV B) represented the faculties of ten schools selected from a list of area schools that did not utilize the environmental education coordinator. Instruments were distributed and returned by mail. Twenty-three usable instruments were returned from subgroup IV A. Twenty—eight usable instruments were returned from subgroup IV B. No follow—up of non— respondents was attempted. Results of data analysis utilized to explore research questions and hypotheses posed for data collection from groups I-IV are presented below. 81 Research Question 1 Can evidence of reliability and validity (content and construct) be established for an instrument developed to measure perceived L of C toward taking specific categories of environmental action in a stated situation? Cronback's alpha reliability coefficients (a measure of internal consistency) were calculated for the total PECM and for each of the PECM subscales. In addition, these coefficients were calculated for data collected from each sample group. Table 4.5 presents the alpha reliability coefficients for total PECM scores within each sample group. Coefficients for each group were above the .70 alpha recommended by Nunnally (1978). This would indicate that the measure taken as a whole exhibits evidence of high internal consistency (reliability). Alpha reliability coefficients for total I, P, and C subscales per group are presented in Table 4.6. Across all sample groups and for each subscale acceptable alpha reliability coefficients were achieved (alpha> .70). These findings support evidence of high internal consistency for the I, P, and C subscales. Table 4.7 presents alpha reliability coefficients determined for total action category subscales within each sample group. Again, all reliability coefficients for respective measures within each sample group were above a .70 alpha indicating that each of the five subscales exhibits evidence of high internal consistency. Content validity of the PECM is based upon planned development of item content and upon a systematic selection of homogeneous items (see discussion on validity Chapter 3). 82 Table 4.5 Reliability Coefficients of Total PECM Scores for each of the Sample Groups. I Variable *Total PECM Group ‘N Alpha I Env. Issues Class 73 .97054 II Bio. Class Group II A 38 .95791 1 (Solid Waste , Issue) Group II B 37 .96075 (Acid Rain Issue) III Sierra Club 10 .98057 IV Teachers Group IV A 21 .98778 (EE Group) Group IV B 26 .97298 (Random Sample) * Total items = 45 83 Table 4.6. Reliability Coefficients of Total I, P and C Subscales for each of the Sample Groups. Variables +Approx. *Internal (I) *Powerful Others (P) *Chance (C) Group ”.N Alpha Alpha Alpha I Env. Issues Class 80 .91231 .93652 .92456 II Bio. Class Group IIA 38 .88252 .90099 .90888 (Solid Waste Issue) Group IIB 38 .88284 .89641 .92865 (Acid Rain Issue) III Sierra Club 10 .93343 .95259 .95394 IV Teachers Group BIA 23 .96652 .96231 .97333 (EE Group) Group B7B 27 .91485 .93101 .92596 (Random Sample) * Total items = 9 per variable + Missing data caused N to vary Slightly for some subscales 84 AoHaEmm Eowcmmv mqwmm. mmqmm. mecca. momma. cemmm. mm m >H macho Aaoouo mmv homom. mecca. mqqmm. ommom. owwmm. mm < >H nacho mnmnommw >H HqNom. memom. mooom. ONOmm. amoom. OH nsao mquHm HHH AmsmmH :Hmm nfiu:m H gym Alfie's» Ham... a mag m a ammoge OHEocoomx ucoaowmcmEoom« o>fimmsmuom« HmOHuHHomR .xouamm moanmfium> .mmaouw mHQEmm emu mo Sumo How CHmOmasm hnowoumo :Ofiuo< HmucmEcOufi>am some mo mCGOfiOHmwmoo %uHHflamHHoM .m.q mHAMH 85 Table 4.8 presents the corrected item total and respective item subscale correlation coefficients for the environmental issues class (group I). Generally, each item tends to correlate rather well with the total scale and/or with.their respective subscales. These results seem to provide evidence that items are functioning in the manner pre- scribed by the objectives posed for item content development. Results of item analysis for data collected from the other groups did not vary substantially from the results reported for group I and therefore are not reported in tabular form. Field study research questions two through six, along with corres— ponding hypotheses, are based on L of C/I,P,C theory and were tested to determine if the PECM exhibits evidence of construct validity. In order to clarify the analysis used to test the proposed hypothesis, descriptive statistics for each of the sample groups are presented in Tables 4.9 through 4.18. Table 4.9 presents descriptive statistics for each sample group based on total PECM scores. All group means are located in the lower half (internal portion) of the possible range of scores (i.e., 45-270). A comparison of means to medians indicates that total PECM scores seem to approximate a normal distribution. On a relative basis (comparison of sample groups), group III (Sierra Club) achieved the lowest mean score identifying it as the most internal of the groups. Group IV A (teachers with EB exposure) achieved the highest mean score identifying it as the most external of the groups. Descriptive statistics for total I, P, and C scores per sample group are presented in Tables 4.10, 4.11, 4.12. All group means for each 86 I I I mommm. I Hmmwm. I I mwmmo. OH I I I mwmqm. I I mhmmo. I qmmn. mg I I I mMNmm. I I I Nomqn. NMHNM. «a I I I qNNmN. I I memes. I mmmqm. mH I I I anhmo. I l I MONON. mmmmo. NH l I I mmomm. I I oowmm. I mNNao. AH I I I mmmom. I MNmNm. I I m~¢wq. OH I I I I madam. mommo. I I Ommqo. m l I I I Roman. l mquN. I mmoqm. m I I I I onmmn. l I meHm. mammo. n I I I I endow. I mfiomn. I Nmfiom. o I I I I mmmmo. Ommmm. I I emwfiw. m I I I I Amoco. I I afifific. «mace. n I I I I O¢HMN. I Hoofim. I Nowdn. m I I I I omOQN. I I «Nmmo. mwNmQ. N l I I I mxmmm. «omen. I I waON. H M. M. m. m. m. m. M. m. M. .02 EouH mamom mHmom mfimom mamom madam mamom madam mamom mamom Hmwod aflfiocoom quEmwmcmEoum o>flmm=mumm HNUHufiHom O m H 20mm Haney .AH nsouov mmmHu wmommH Hmuaoacoufi>am osu How mcowumHouuoo manomnsm EmuH monomuuoo o>wuuommmm new Hmuoa EmuH emuoouuoo mo xwuunz .w.q mHan 87 I ammom. I I I qwmmo. I I «cemm. cm I mmmmm. I I I I mmNON. I mmmmo. mm I oqum. I I I I I NmMON. mmwwm. mm I oqum. I I I waom. I I oomom. Hm I mmHnm. I I I I I onmo. Nmmwm. cm I mNmON. I I I I ommmc. I oqumo. am I oocmm. I I I owmmq. I I ecmwm. mm I I Hummo. I I I HHqu. I «mmmm. mm I I «mon. I I qowoo. I I “ammo. em I I ammon. I I I mqmqo. I quoo. mm I I emcee. I I I I «mamm. «mamm. «N I I «memo. I I I mquo. I oomwm.. mm I I «meme. 1 I Hmmmm. I I mmomo. mm I I NNme. I I I I oqmmm. chmm. Hm I I NHwHN. I I mnemo. I I «mace. ON I I «moms. I I I I Hmmmm. mmomq. mH I I I cmman. I mnmmo. I I moowo. mH I I I cemmo. I _ I I mommm. mmmac. NH M. M. M. m. M. m. m. m. m. .02 EmuH mHmom mHmom mHmom onom OHmom OHmum mHmom mHmom OHmom Hmwmq OHEosoom ucoamwmcmEoom o>Hmm=mumm HmOHuHHom o m H 20mm sauce .He.oeooV .m.s manna 88 fimqow. I I I I I mqwoo. I momwo. mq quqm. I I I I I mownm. I Hence. qq mmomw. I I I I wmofio. I I mmomo. mc «mann. I I I I I omqme. I mofioo. Nq mqnmm. I I I I I wmmmm. I omwom. Hq Hmonm. I I I I I mmmqo. I mmqmo. oq momfim. I I I I oqmqm. I I 00mfim. mm mwomo. I I I I I I qumq. mmmmm. mm qumw. I I I I qmoao. I I mm¢o~. um I qumw. I I I I mwmoo. I macaw. on I mfifimm. I I I I I nmfiwm. NNnmm. mm m. m. m. m. m. m. M. m. m. .02 EmuH «Hmom mfimom meum madam mamum mamum mamom mamom mamom Hmwmq UHEocoum ucmawwmcmaoum w>wmm3mumm HmUHuHHom U m H 20mm Hauoa 2:283 .mé $33 89 m.fimfi “afloawflz ONNImq wmuoom mo mmcmm mfinwmmom« Afiucv Afiucv mao.~q cem.mfifi mofi.mfifi cm mmfi 0N m >H msoyo Afiusv Amuav A.um .>amv who.mm oo.qmfi mmo.oq~ mq «ow MN < >H aaogo mum:u¢m9 >H Aflucv Afincv ofio.om oom.mm o05.mm em amfi OH nafio muuofim HHH Aducv Aguav Acflmm ufio:m H mm v2 2 wuoom muoum z macho so; cw“: mwcmm* meEmm 50mm How mmuoom 20mm HauoH mo mSOHumH>mQ wumUCMum vcm mcmfivmz .mcmwz .mmuoom .maouo mo mwcmm .m.q maan 9O m.~m ucfiomufiz oaImH mmHoum we mmcmm manflmmom« AHHGV AHHGV Awfimamm Eowcmmv Hom.- mmm.mm www.mm ma mm mm m >H macaw Afiusv Amuav A.wm .>:mv mmqi 80.3 53.3 2 cm 2 ¢ 5 965 mumnumma >H :15 2ch Hmw.m ooo.~m ooo.mm ma mm OH nuao mpuoflm HHH . Afiucv Amusv flawmm vHucm H mm fix 2 wuoom muoom z msouo 33 53m mwcmm« .msouw mamfimm :umw How mmhoum fiHV HmcumuCH mo macaumfi>wo wumwcmum wan mcmfiwmz .mcmmz .mwnoom mo omcmm .o~.q magma 91 9% 283m: omImfi mmuoom mo mwcmm mafiamm0h¢ Amucv ANHCV AmHaEmm Eowcmmv mNo.oH oom.qq Hmo.mq NH mm mm m >H macho Sufi SHE 63 5:5 omm.wH ooo.oq mo~.wq ma qw Hm ¢ >H asouo wuwnomme >H 3ch SHE ww~.- oom.mm ooq.mm ON mm o~ asao muumflm HHH ans :3 95 Bus qu.mH ooo.oq wNH.N¢ ma mm mm m HH asouw Sufi :ué A383 338 oom.- omm.~< mmm.~¢ ma No mm < HH msouo mmmao .owm HH SIB Sufi mama Nmo.qH ooo.Hq HNN.Nq ma ca RN mmsmmH .>cm H mm w: 2 muoom muoow z asouo 33 :3: wmsmm« .masouu mHQEMm comm wow mwuoom Amv muonuo stumzom Hmuoa mo mGOHuMfl>wo wumwamum wad mcmfiwoz .mcmoz .wmuoum mo mmwcmm .-.¢ wanmh a.mm uafloaufiz omImH mmuoom mo wwcmm anHmmom* Awndv Afincv mmHaEMm Eowcmmv wmm.q~ ooo.Hq wa.oq wfi we m >H asouo Afiuav ANucV A.um .>cmv “wo.mfi ooo.~¢ omfi.oq ma cm mN < >H aaogo muwfiumme >H “Hucv “aucv Noo.ofi ooo.Hm oom.Hm ma we OH asfio mgumfim HHH 2 o, Afiucv Hfiuav Acfimm ufio:m H Om $2 2 muoom ouoom z asouo 30g swfi: mwcmm« .msouw wHQEMm 50mm uQm mmuoum ADV wucmno Hmuoa mo macaumfi>uo vumwamum vcm wcmfiwmz .mcwmz .mmuoom mo wmcmm .NH.¢ magma 93 subscale are located in the lower half (internal portion) of the possible range of scores (15—90). Based on a mean and median comparison scores seem to approximate a normal distribution. On a relative basis, group III scored lowest (most internal) and group IV A scored the highest (most external) on each of the three sub- scales. In comparing I, P, C scores within each sample group, no substantial differences seem apparent between I and C scores. However, P scores are consistently higher than I or C scores for each of the sample groups. Tables 4.13 through 4.17 present descriptive statistics for each ‘ sample group based on total environmental action category scores. All group means for each subscale are located in the lower half (internal portion) of the possible range of scores (i.e., 9—54).I Subscale scores for each of the groups also seem to approximate normal distributions. The Sierra Club (Group III) scored the most internal across all subscales and group IV A (teachers with EE exposure) scored most external on all five subscales. Another notable trend was for all groups to generally score legal action lower (more internal) than other actions. Table 4.18 presents the means and standard deviations of citizen action scores for each group. Members of the Sierra Club (Group III) reported more action taking than other groups. The environmental issues class reported the least amount of citizen action taking. Research Question 2 What relationship exists between scores on a measure of citizen action and total PECM scores within the environmental issues class, the Sierra Club group, each of the teachers subgroups and each of the introductory biology subgroups? 94 m.Hm “chosen: qmla monoum mo mwcmm wanflmmomx ANHEV AHHSV Aofiaamm Eowcmmv new.m omN.qN 0mm.eN m we RN m >H macho fiance xNucv ~.em .>:mv wa.zz coo.eN qez.wN m mm mm < >H macho mumnuwwa >H Amucv flflucv woz.o ooo.fiz ooq.m~ ma mm as nsflo mnemflm HHH Amuse Amuse Assam enu:m H mm w: E whoom muoom z msouw 3o; swam wwcmm* .maouo manfimm some now mwuoom 50Huo< w>Hmm=mHmm Hmuoe mo mfiOfiumw>wo wumvcmum can mamapoz .mcmmz .mwuoom mo mwcmm .mH.q oaame 95 m.zm “chosen: qum mmuoom mo mwcmm mmmwmmoma Afiudv AHHCV AmHQEmm Eovcmmv Nem.0z oom.¢~ om~.o~ m we mm m >H nacho Afiucv Amncv A.wm .>:mv omo.zz ooo.m~ ooo.NN m as mm a >H anouo mumsume >H Azucv Amuse mom.e oom.ma ooo.wz on mm oi nafio «spasm HHH Amflflv AHHEV Adwmm Uflocm H mm n: ma muoom muoom z macho Bog emu: |IIS|I|I| mafiadx mmuoom coauom .msouu onEmm comm now Hmofluflaom Hauow mo macaumfi>ma pumwcmum mam mcmfivwz .mcmmz .wwuoom mo mwcmm .q~.q maan 96 Amucv AHHCV Amamemm Bowcmmv enm.oz ooo.a~ seq.m~ a me mm m >H scope A.em .>cmv see.Oz om~.o~ qom.w~ m cm mN < >H macho mumnomme >H fiance Aauav mmH.w oom.m~ ooo.m~ m mm Oz gaze muumnm HHH Awncv Amuse Achm eflo:m H mm fix 2 wuoom wuoow z asouu seq swam owCMMx coauo< unmawwmamaoom HouOH mo mEOHuma>wa pumpcMum was mcmwumz .mcmoz .mmuoom mo mwcmm .asouo meEmm comm you mmuoom .m~.¢~ anmH m . a m “£33: qum mmuoom mo mwcmm wahwmmomx Amusv AHucv Amaaamm aovcmmv mmo.ofi 00m.HN qwm.~m m me mm m >H asouo Agnew Amucv A.wm .>cmv 20.2 08.8 892 a em 2.. < E 955 muwcommh >H Snav SHE qmo.m coo.mH oom.om ofi mm oH gnaw mnuwflm HHH W €an 305 mNH.HH NoH.NN , oom.¢m m mm oe m HH anouo Amncv AHHGV Amummz Uflaomv 03A 892 898 m mm mm < 3 uses mmmao .oam HH Aoucv Aancv mmmau Omm.m m~m.mm Hmm.mm m cm mm mmsmmH .>:m H mm w: z ouoom muoom z macho 33 amen wmcmm« .asowo meEmm Sumo now mmuoom cowuo< owEocoom Hmuoe mo mefiumH>mo vnmwSMum was mcmfiwwz .wcmwz .mmuoom mo mwcmm .o~.¢ MHAMH 98 m.Hm ucHomeHz wmIo mmuoom wo wwcmm mHnHmmom¥ Amucv AHucv AmHQEMm Eowcmmv qu.oH oom.HN www.mw m Hm mm m >H moose Awncv Amuse A.em .>:mv HHs.NH ooo.c~ Hmw.m~ a em HN < >H nacho mumnummk >H AHucv Aanv omo.H cem.wH oom.mH oH Hm oH nsHo muumHm HHH Aquav AHucv Achm eHocm H mm U2 2 muoom whoom z macho 30H emHm wwcme .asouo oaaemm comm HOw mwgoom cOHuu< flowed Houoe mo mCOHudH>wQ pumwcmum cam mcmemz .mcmmz .mmpoom mo omcmm .mfi.q wanna 99 4.18. Means and Standard Deviations of Citizen Action Scores for Each Sample GrOup. Group I M £12 I Env. Issues 81 13.827 11.175 Class II Bio. Class II A 38 14.394 11.068 (Solid Waste) II B 40 15.850 13.284 (Acid Rain) J III Sierra Club 10 44.900 15.242 IV Teachers IV A 28 15.087 11.036 (Env. Ed.) IV B 28 24.6429 16.5981 (Random Sample) Possible Range of Scores 0—121 Research Hypothesis 2.1. A significant (p.: .05) negative correlation will exist between scores on a measure of citizen action and total PECM scores within each of the selected groups; Correlational statistics were utilized to investigate research questions two through five. A Pearson's r was calculated for the variables iden- tified in each question. Rationale for use of the Pearson's r is the same as that provided in the discussion of pilot study II analysis. Results of the Pearson's r between total PECM scores and citizen action scores for each sample group are presented in Table 4.19. With L the exception of groups II B and IV B, significant negative correlations (p.: .05) were obtained. These significant correlations support the proposed hypothesis (2.1) and indicate that, for each group, as total PECM scores increase citizen actions reported decrease. Groups II B and IV B did not yield significant results although the correlations are in the hypothesized direction. In summary, total PECM scores for each group tend to correlate negatively with citizen action scores and the majority (4 out of 6) of correlations were significant. With the exception of subgroup II B, all groups were administered the PECM with the solid waste issue summary. As a further check on con- struct validity data from the five groups were collapsed and a Pearson's r was calculated between total PECM scores and citizen action scores (Table 4.19). Again, a significant negative correlation was achieved. Research Question 3 What relationship exists between scores on a measure of citizen action taking and total I, P, and C scores for the environmental issues 101 Table 4.19; Pearson Product-Moment Correlations for Total PECM Scores with Citizen Action Scores for each of the Sample Groups and for Sample Groups Responding to the Solid Waste Issue Combined. Total PECM Scores/Citizen Action Scores Correlation Significance Sample Coefficient Level Size Group 5 B E I Env. Issues —.3391 .001* 81 Class II Bio. Class I Group II A —.5958 .001* 38 (Solid Waste) Group II B 0.2086 .098 40 (Acid Rain) III Sierra Club -.6320 .025* 10 IV Teachers Group IV A -.4843 .013* ’ 21 (Env. Ed.) Group IV B -.2318 .127 26 (Random Sample) All Groups Combined —.3451 .001* 168 (Except II B) * Significant relationship (P j_.05) 102 class, the Sierra Club group, each of the teachers subgroups and each introductory biology subgroup? Research Hypotheses: 3.1. A significant negative correlation (p_: .05) will exist between scores on a measure of citizen action and total I scores for each selected group. 3.2. A significant negative correlation (p.: .05) will exist between scores on a measure of citizen action and total P scores for each selected group. ‘ 3.3. A significant negative correlation (p §_.05) will exist between a measure of citizen action and total C scores for each selected group. Results of Pearson's r correlations between citizen action scores and I, P, C scores are presented in Tables 4.20 through 4.22. With the exception of group II B and IV B, significant negative correlations were obtained between I, P, and C scores and citizen action scores. Significant findings support research hypotheses 3.1, 3.2 and 3.3 and indicate that as I, P, or C scores increase, citizen action reporting decreases. A significant negative correlation was determined for group II B on the C subscale but not on the I and P subscales. No significant correlations were achieved for groups IV B on the I, P, and C subscales. However, nonsignificant correlations for groups II B and IV B were all in the hypothesized direction. Thus, all I, P, C subscale scores for each sample group correlated negatively with reported citizen action taking. 103 Table 4.20. Pearson Product-Moment Correlation for Total Internal (I) Scores with Citizen Action Scores for each of the Sample Groups. Internal (I) Scores/Citizen Action Scores Correlation Significance Sample Coefficient Level Size Group _£ .2 3 I Env. Issues —.3614 .001* 80 Class II Bio. Class Group II A —.5927 .001* 38 ‘ (Solid Waste) 4 Group II B —.2673 .052 38 (Acid Rain) III Sierra Club —.6361 .024* 10 IV Teachers Group IV A —.4908 .009* 23 (Env. Ed.) Group IV B —.2925 .069 27 (Random Sample) 9 * Significant relationship (p :_.05) 104 Table 4.21. Pearson Product—Moment Correlation for Total Chance (C) Scores with Citizen Action Scores for each of the Sample Groups. Chance (C) Scores/Citizen Action Scores Correlation Significance Sample Coefficient Level Size Group _r_ p 2 I Env. Issues -.2827 .006* 78 Class II Bio. Class Group II A .5447 .001* 38 I (Solid Waste) ' Group II B -.3223 .023* 39 (Acid Rain III Sierra Club -.5650 .024* 10 IV Teachers Group IV A -.4777 .011* 23 (Env. Ed.) Group IV B —.2808 .078 27 (Random Sample) * Significant relationship (p i .05) 105 Table 4.22. Pearson Product-Moment Correlation for Total Powerful Others (P) Scores with Citizen Action Scores for each of the Sample Groups. Powerful Others (P) Scores/Citizen Action Scores Correlation Significance Sample Coefficient Level Size Group E. R E I Env. Issues -.2719 .008* 77 Class II Bio. Class Group II A —.5238 .001* 38 i (Solid Waste) ' Group II B -.2o94 .100 39 (Acid Rain) III Sierra Club -.6358 .024* 10 IV Teachers Group IV A -.3984 .037* 21 (Env. Ed.) Group IV B -.1361 .245 28 (Random Sample) *Significant relationship (p 5_.05) 106 Research Question 4 What relationship exists between scores on a measure of citizen action and total I+P+C scores for each environmental action category within the environmental issues class, the Sierra Club group, each of the teachers subgroups, and each of the introductory biology sub- groups? Research Hypothesis: 4.1. A significant negative (p.: .05) correlation will exist be- tween scores on a measure of citizen action and total I+P+C scores for each environmental action category within each of the selected groups. Results of Pearson's r correlations between total environmental action category scores and citizen action scores are presented in Tables 4.23 through 4.27. Significant coefficients (p :_.05) support hypothesis 4.1 and indicate that as the respective environmental action category score increases the amount of reported citizen action decrease. The ecomanagement subscale (Table 4.25) provided the fewest number of significant negative correlations (3 out of 6) across groups. Al- ternatively, the persuasive action (Table 4.23) subscale provided the largest number of significant negative correlations (5 out of 6). Although analysis of some group data did not yield significant results for a particular subscale, overall correlations were in the hypothesized direction. 107 Table 4.23. Pearson Product-Moment Correlations for Total Persuasive Action Scores with Citizen Action Scores for each of the Sample Groups. Total Persuasive Action Scores/Citizen Action Scores Correlation Significance Sample Coefficient Level Size Group E B E I Env. Issues —.3038 .003* 79 Class II Bio. Class Group II A -.3996 .001* 38 i (Solid Waste) ; Group II B —.3090 .026* 40 (Acid Rain) III Sierra Club -.7027 .012* 10 IV Teachers Group IV A -.3929 .032* 23 (Env. Ed.) Group IV B -.1518 .225 27 (Random Sample) * Significant relationship (p :_.05) 108 Table 4.24. Pearson Product-Moment Correlations for Total Political Action Scores with Citizen Action Scores for each of the Sample Groups. Total Political Action Scores/Citizen Action Scores Correlation Significance Sample Coefficient Level Size Group _r_ p E I Env. Issues -.3489 .001* 81 Class II Bio. Class Group II A -.4244 .004* 38 1 (Solid Waste) Group II B —.2982 .031* 40 (Acid Rain) III Sierra Club -.7576 .006* 10 IV Teachers Group IV A —.3393 .057 23 (Env. Ed.) Group IV B —.0944 .316 28 (Random Sample) * Significant relationship (p §_.05) 109 Table 4.25. Pearson Product—Moment Correlations for Total Ecomanagement Action Scores with Citizen Action Scores for each of the Sample Groups. Total Ecomanagement Action Scores/Citizen Action Scores Correlation Significance Sample Coefficient Level Size Group E p E I Env. Issues —.1702 .1702 78 Class II Bio. Class 3 Group II A —.3439 .017* 38 '” (Solid Waste) Group II B -.1390 .199 39 (Acid Rain) III Sierra Club —.1926 .297 10 IV Teachers Group IV A —.4280 .021* 23 (Env. Ed.) Group IV B -.3968 .020* 27 (Random Sample) * Significant relationship (p‘: .05) 110 Table 4.26. Pearson Product-Moment Correlations for Total Economic Action Scores with Citizen Action Scores for each of the Sample Groups. Total Economic Action Scores/Citizen Action Scores Correlation Significance Sample Coefficient Level I Size Group .E .2 E I Env. Issues —.2644 .009* 79 Class II Bio. Class . I Group II A —.5240 .001* 38 K (Solid Waste) Group II B —.2658 .049* 40 (Acid Rain) III Sierra Club —.4789 .081 10 IV Teachers Group IV A -.4478 .016* 21 (Env. Ed.) Group IV B -.2730 .080 . 28 (Random Sample) * Significant relationship (p j .05) 111 Table 4.27. Pearson Product-Moment Correlations for Total Legal Action Scores with Citizen Action Scores for each of the Sample Groups. Total Legal Action Scores/Citizen Action Scores Correlation Significance Sample Coefficient Level Size Group E p n I Env. Issues -.2158 .027* 80 Class II Bio. Class Group II A -.4760 .001* 38 (Solid Waste) Group II B -.1856 .132 38 (Acid Rain) III Sierra Club -.6052 .032* 10 IV Teachers Group IV A -.5305 .007* 21 (Env. Ed.) Group IV B -.0520 .396 28 (Random Sample) * Significant relationship (p: .05) 112 Research Hypothesis: 5.1. A significant positive (p <.05) correlation will exist between total I, P, and C scores within each of the groups selected (a relatively high I score refers to lack of belief in internal control). A Pearson's r was used to calculate correlations between I, P, and C scores within each of the sample groups (Table 4.28). Significant positive correlations, which support hypothesis 5.1, were determined between I, P, and C scores within each of the sample groups. To interpret these results clearly it should be reiterated that I item scoring is reversed from P and C item scoring. A relatively high score on the I items will reflect a low belief in that orientation (external control). A relatively high score on the P or C items will reflect a high belief in those orientations. Alternatively, relatively low scores on the I items will reflect a high belief in that L of C orientation; whereas a relatively low score on P and C items reflects a lack of belief in those orientations. Thus, correlations in Table 4.28 indicate that as respondents agree or disagree with the P orientation, they respectively agree or disagree with the C orientation. However, as respondents agree with the I orientation (personal control), they tend to disagree with P and C orientations. Research.Question 6 Is there a significant difference between total I, P, or C scores within the environmental issues group, the Sierra Club group, each of the teachers subgroups and each introductory biology subgroup? Table 4.28. Pearson Product-Moment Correlation for Total I, P, and C Scores within each Sample Group. Variables Correlated P/C P/I C/I Group _E _r _r I Env. Issues .8502* .8068* .8187* Class n=74 n=76 n=77 II Bio. Class Group II A .7661* .7108* .8747* (Solid Waste) n=38 n=38 n=38 Group II B .8559* .7461* .7886* (Acid Rain) n=38 n=38 n=37 III Sierra Club .8994* .9131* .9079* n=10 n=10 n=10 IV Teachers Group IV A .9521* .8609* .8102* (Env. Ed.) n=21 n=21 n=23 Group IV B .9400* .7532* .8125* (Random Sample) n=27 n=27 * Significant relationship (P j_ .05) 114 Research Hypothesis: 6.1. A significant difference (p_: .05) in mean I, P, and C scores will be determined within sample groups. Research questions 6 and 7 were posed to investigate the possibility of significant differences between selected subscale scores, by sex within each sample group. The SPSS MANOVA Profile Analysis program was employed for analyses. Results in Tables 4.29 through 4.34 denote that within each sample group there was no indication of significant interactions between sex and I, P, C subscales. In addition, with the exception of the environ- mental issues class, no significant (p-: .05) difference in response by sex was found. A significant difference in reaponse by sex was discovered for the environmental issues class (Table 4.29). Based on a breakdown by sex of mean total PECM scores (males: M=129.700, Females: M299.857) and on a lack of interaction between sex and measures it can be assumed that females scored significantly lower (internal) on all three measures than did males of the same group. To determine differences between measures, I, P, and C means for each group were rank ordered (high to low). Results for the environ- mental issues class (Table 4.29) indicate that P scores were significantly higher (p=.00002) than C scores and I scores did not differ significantly from C scores at the .05 level. Because mean scores were rank ordered, the following assumption for P vs I scores can logically be applied; If P>C and C=I, then it is assumed P>I. In summary, within female and 115 Table 4.29. Repeated Measures Analysis of Variance for I, P, C Scores by Sex for the Environmental Issues Class (Group I). TOTAL PECM Sex N M SD Male 40 129.700 37.799 Female 28 99.857 29.431 ANOVA Summary Multivariate Univariate Variables df F F P (approx. F) Interaction between 2, 65 .3089 .73534 sex and measures Difference by sex 1, 66 12.2378 .00084* Difference between (approx. F) measures 2, 66 13.9000 .00001* + P vs C 1, 67 21.44 .00002* + C vs I l, 67 .6063 .43892 P vs I Assumption: If P>C and C=I, then it is assumed P>I * Significant relationship (p:.05) + Means were rank ordered from high to low prior to analysis 116 Table 4.30. Repeated Measures Analysis of Variance for I, P, C Scores by Sex for the Biology Group (II A) Assigned the Solid Waste Issue. TOTAL PECM Sex N M SD Male 6 104.166 32.811 Female 32 113.093 27.793 ANOVA Summary Multivariate Univariate 4, Variables df F F P Interaction between (approx. F) sex and measures 2, 35 .5994 .55467 Difference by sex 1, 36 .4942 .48657 Difference between (approx. F) measures 2, 36 13.876 .00003* + P vs C 1, 37 21.145 .00005* C vs I 1, 37 2.441 .12673 P vs I Assumption: If P>C and C=I, then it is assumed P>I. * Significant relationship (p_: .05) + Means were rank ordered from high to low prior to analysis. 117 Table 4.31. Repeated Measures Analysis of Variance for I, P, C Scores by Sex for the Biology Group (II B) Assigned the Acid Rain Issue. Total PECM Sex N M SD Male 10 120.100 53.175 Female 26 122.346 28.168 ANOVA Summary Multivariate Univariate Variables df F F P Interaction between (approx. F) sex and measures 2, 33 .7313 .48894 Difference by sex 1, 34 .0303 .86294 Difference between (approx. F) measures 2, 34 6.1569 .00522* + P vs I 1, 35 .53277 .47030 I vs C 1, 35 4.20623 .04782* P vs C Assumption: If P=I and I>C, then it is assumed P>C. * Significant relationship (p_: .05) + Means were rank ordered from.high to low prior to analysis. 118 Table 4.32. Repeated Measures Analysis of Variance for I, P, C Scores by Sex for the Sierra Club Membership (Group (III). Total PECM Sex N M SD Male 7 97.571 37.216 Female 3 104.661 15.176 ANOVA Summary Multivariate Univariate Variables df F F P Interaction between (approx. F) sex and measures 2, 7 .7047 .52621 Difference by sex 1, 8 .1015 .75817 Difference between (approx. F) measures 2, 8 3.12621 .09927 + P vs I 1, 9 2.781 .12973 I vs C 1, 9 1.444 .26031 P vs C Assumption: If P=I and I=C then it is assumed P=C. + Means were rank ordered from high to low prior to analysis. [—v—H—H. Table 4.33. Repeated Measures Analysis of Variance for I, P, C Scores by Sex for the Teachers Group Utilizing the Environmental 119 Education Consultant (Group IV A). Total PECM Scores Sex N M SD Male 4 123.500 61.235 Female 17 144.000 55.556 ANOVA Summary Multivariate Univariate Variables df F F P Interaction between (approx. F) sex and measures 2, 18 .8262 .45364 Difference by sex 1, 19 .4264 .52157 Difference between (approx. F) measures 2, 19 1.14575 .33900 + P vs C 1, 20 1.65919 .21242 C vs I l, 20 .20581 .65495 P vs I Assumption: If P=C and C=I, then it is assumed P=I. + Means were rank ordered from high to low prior to analysis. 120 Table 4.34. Repeated Measures Analysis of Variance for I, P, C Scores by Sex for the Teachers Subgroup not Utilizing the District Environmental Education Consultant (Group IV B). Total PECM Sex N M SD Male 6 132.500 55.207 Female 19 116.315 ANOVA Summary Multivariate Univariate -h Variables df F F P Interaction between (approx. F) sex and measures 2, 22 1.0472 .36776 Difference by sex 1, 23 .6716 .42091 Difference between (approx. F) measures 2, 23 8.7299 .00151* + P vs C 1, 24 11.911 .0028* C vs I 1, 24 5.848 .02355* P vs I Assumption: if P>C and C>I, then it is assumed P>I. * Significant relationship (p‘: .05) + Means were rank ordered from high to low prior to analysis. 121 male groups of the environmental issues class P scores were significantly greater than I and C scores. However, I, P, and C scores were signi- ficantly lower for females as compared to males. Excluding differences by sex, the biology class exposed to the PECM containing the solid waste issue displayed results similar to that of the environmental issues class (Table 4.30). P scores were significantly greater (p=.00005) than C scores and again I scores did not differ significantly from C scores. It was assumed P scores were greater than I scores. Data analysis for the biology group exposed to the acid rain issue i (Table 4.30) indicates that P scores were not significantly different from I scores. However, I scores were significantly higher than C scores so it was logically assumed that P scores were significantly higher than C scores. Analysis of data for the Sierra Club (group III) and the teachers group utilizing the enivronmental education consultant yielded no significant differences at the .05 level between the respective sub- scales. For the teacher group not utilizing the environmental education consultant (group IV B) significant differences between all three measures was found (Table 4.34). In general findings support hypothesis 6.1 because significant differences were found between I, P, and C scores within select groups. Research Question 7 Is there a significant difference (p‘: .05), by sex, between total environmental action category scores within the environmental issue 122 group, the Sierra Club group, each teacher subgroup and each of the introductory biology subgroups? Tables 4.35 through 4.40 present the results of a profile analysis program calculated for each sample group. Results indicate there were no interactions between sex and measures within any of the groups. In addition, with the exception of the environmental issues class, no group exhibits a difference by sex of a difference between measures. Data from the environmental issues group produced both a difference by sex and a difference between measures (Table 4.35). Based on a breakdown by sex of mean total PECM scores (males; M—129.700, Females; M-99.857) and on the lack of interaction between sex and measures, it can be assumed that females scored significantly lower (i.e., more internal) on all five subscales. Subscale scores were again rank ordered (high to low) to determine specific differences between measures. Results indicate that legal scores were significantly lower (p=.00329) than other scores. No other measures differed significantly from one another. Thus, within female and male groups of the environmental issues class, legal scores were significantly lower than the other four environmental action category scores. However, all environmental action category scores were significantly lower for females as cempared to males. Research Question 8 Do biology subgroups differ significantly (p_: .05) on total PECM 'scores, I, P, C scores, total environmental action category scores and/or citizen action scores? 123 Table 4.35. Repeated Measures Analysis of Variance for Total Action Category Scores by Sex for the Environmental Issues Class (Group I). Total PECM Sex N M SD Male 40 129.700 37.799 Female 28 99.857 29.431 ANOVA Summary Multivariate Univariate 1, Variables df F F P Interaction between sex and measures Difference by sex Difference between measures +Econ. vs Ecomgmt. Ecomgmt. vs P01. P01. vs Persuasive Persuasive vs Legal (approx. F) 4, 63 .8856 .47784 1, 66 12.2378 .00084* (approx. F) 4, 64 5.8397 .00045* 1, 67 .5456 .46269 1, 67 .7780 .38090 1, 67 .2283 .63431 1, 67 9.29636 .00329* * Significant relationship (p_: .05) + Means were rank ordered from high to low prior to analysis. 124 Table 4.36. Repeated Measures Analysis of Variance for Total Action Category Scores by Sex for the Biology Class (Group II A) Assigned the Solid Waste Issue. Total PECM Sex N M SD Male 6 104.166 32.811 Female 32 113.093 27.793 ANOVA Summary Multivariate Univariate Variables df F F P Interaction between (approx.FU , sex and measures 4, 33 2.25392 .08439 Difference by sex 1, 36 .49424 .48657 Difference between (approx. F) measures 4, 34 .74665 .57607 125 Table 4.37. Repeated Measures of Analysis of Variance for Total Action Category Scores by Sex for the Biology Class (Group II B) Assigned the Acid Rain Issue. Total PECM Sex N M SD Male 10 120.10 53.175 Female 26 122.346 28.168 ANOVA Summary Multivariate Univariate Variables , df F F P Interaction between (approx. F) sex and measures 4, 31 1.8416 .14598 Difference by sex 1, 34 .0303 .86294 Difference between (approx. F) measures 4, 24 .7372 .57351 126 Table 4.38. Repeated Measures Analysis of Variance for Total Action Category Scores by Sex for the Sierra Club Membership (Group III). Total PECM Sex N M SD Male 7 97.571 36.216 Female 3 104.661 15.176 ANOVA Summary Multivariate Univariate Variables df F F P Interaction between (approx. F) sex and measures 4, 5 3.8371 .08637 Difference by Sex 1, 8 .1015 .75817 Difference between (approx. F) measures 4, 6 1.1510 .41711 127 Table 4.39. Repeated Measures Analysis of Variance for Total Action Category Scores by Sex for the Teacher Group Utilizing the Environmental Education Consultant (Group IV A). Total PECM Sex N M SD Male 4 123.500 61.235 Female 17 144.000 55.556 ANOVA Summary Multivariate Univariate Variables df F F P Interaction between (approx. F) sex and measures 4, 16 2.17761 .11803 Difference by sex 1, 19 .42642 Difference between (approx. F) measures 4, 17 2.2097 .11128 Table 4.40. 128 Repeated Measures Analysis of Variance for Total Action Category Scores by Sex for the Teachers Group Not Utilizing the Environmental Education Consultant (Group IV B). ' Total PECM Sex N M, SD Male 6 132.500 55.207 Female 19 116.315 37.761 ANOVA Summary Multivariate Univariate Variables df F F P Interaction between (approx. F) sex and measures 4, 20 1.3494 .28658 Difference by sex 1, 23 .61716 .42091 Difference between measures (approx. F) 2.6094 .06485 129 A one-way analysis of variance was calculated to determine if the two biology subgroups differed significantly on total PECM scores. Results in Table 4.41 indicate that no significant differences exist between the two groups at the .05 level. To determine if the two biology groups differed on I, P, C scores, the SPSS MANOVA Profile Analysis program was employed. Results (Table 4.42) indicated a significant interaction between groups and measures. It was further determined that the I subscale was the only subscale that issued a significant differential effect. With respect to differences on total action category scores be— j tween the two groups, it was inferred based on related analysis, that the two subgroups did not differ significantly. Data in Tables 4.35 and 4.36 show that environmental action category scores do not differ significantly from one another within each of the subgroups. In addition, there were no significant differences between the two groups with re— spect to total PECM scores (Table 4.41). As a result of these two findings, it was inferred that the biology subgroups did not differ significantly on total action category scores. Table 4.43 presents the results of a one-way analysis of variance based on citizen action scores. No significant difference between scores on citizen actions was determined at the .05 level. Research Question 9 Do teacher subgroups differ significantly (p §_.05) on total PECM scores, I, P, C scores, environmental action category scores and/or citizen action scores? 130 Table 4.41. One—Way Analysis of Variance Between Biology Subgroups II A and II B with Respect to Total PECM Scores. Total PECM Score Group N M SD Bio. Class II A 38 111.684 28.347 (Solid Waste) Bio. Class II B 37 119.918 37.138 (Acid Rain) ANOVA Source df 53 F P Between 1 1271.213 1.1687 .28317 Within 73 79384.967 131 xfimqfio. NOONN.© mH .H H m> u eoHom. Nmeo.H mH .H o m> H uuoHHm HmHuamumHHHa IMHNHO. I «Namo.e NH .N mausmmma wan masouw Am .xoummmv cmmBumn doauomumucH .m m. mm. moHanHm> mumaum>fica mumHHm>wuH=z hudfififim <>oz< AaHmm eHuav Hoe.mH maH.mm «Hm.mH wNH.~e mNH.~H oom.mm OH H HH mmmHo .on . Hanan: eHHomv Hoe.oH awH.mm oom.HH mam.zq Hmw.m oom.8m mm a HH mmmHo .on elm m mm. m mm. m .m anouo .xouam< doomsu muwnuOIm HmaumuoH .Am HH mam < HH mmaouwnamv mmaouwnam hwoaoam 039 mnu ommBumn mmuoom o .m .H OH oomammm SuHB madmaum> mo mfimhamo< mdudmmmz vmumommm .Nq.q sandy 132 Table 4.43. One—Way Analysis of Variance between Biology Subgroups II A and II B with Respect to Citizen Action Scores. Citizen Action Score Group N M SD Bio. Class II A 38 14.394 11.069 (Solid Waste) Bio. Class II B 40 13.284 13.285 (Acid Rain) ANOVA Source df ss F P Between 1 41.270 .27474 .60169 Within 76 11416.179 133 As part of a SPSS MANOVA Special Contrasts program, a one-way analysis of variance was employed to test for a difference between teacher subgroups with respect to total PECM scores. Results (Table 4.44) indicated no significant difference between subgroups at the .05 level. To determine if teacher subgroups differed on I, P, C scores or environmental action category scores, the SPSS MANOVA Profile Analysis program was employed. Results (Table 4.45) indicated there was no significant interaction between groups and I, P, C measures. In addition, there was no difference between groups with respect to these measures. It was inferred from related analysis, that teacher subgroups did not differ significantly on environmental action category scores. Tables 4.37 and 4.38 indicated that environmental action category scores did not differ from one another within each of the subgroups. In addition, there were no significant differences between the two groups with respect to total PECM scores (Table 4.42). Based on these two findings, it was assumed that teacher subgroups did not differ significantly on total action category scores. To determine differences between teacher subgroups with respect to citizen action scores, an analysis of variance was calculated (i.e., as part of a Special Contrasts program). IResults (Table 4.46) indicated a significant difference between the two subgroups. Thus, the random sample of teachers reported taking more citizen actions than did the teachers utilizing the environmental education consultant. 134 Table 4.44. Analysis of Variance (Planned Contrasts) Based on Total PECM Scores between Sample Groups Administered the Solid Waste Issue Summary. Group N 'M SD I Env. Issues Class 73 117.301 37.401 II Intro. Bio. Class 38 111.684 28.347 (A - Solid Waste) III Sierra Club Members 10 99.700 30.616 IV Teachers (A - EE Teachers) 21 140.095 55.674 (B - Random Sample) 26 . 119.769 41.095 ANOVA Groups 33 df F P Within Cells 243104.105 III vs I, II A, IV A & B 4697.894 1 3.1499 .07780 I, II A vs IV A & B 6493.872 1 4.3541 .03848* I vs II A 285.931 1 .19172 .66202 IV A vs IV B 3748.482 1 2.5733 .11482 * Significant relationship (p i .05) . 135 oaomH. moNHo.N I no .H oooow so ooooHoHHHn mmmmm. I mmmmm. «v .N mandamus mam mmsouw Am .xoummmv coosuon aOHuumHouoH .M .M .M .MW moHnMfiHm> oumHHm>HcD mumHHm>Huasz >Hmaadm <>oz< mmw.qH wa.oq mmo.oH Hmo.mq Hom.HH omm.mm mm muosomow MMIcoz www.mH omH.oq omH.wH mo~.wq mmq.nH mHm.m¢ mm muonomoH mm mm a a. H mm a. a sea .xouaa< mocmso muscuOIm HmououaH .Am >H pom < >H masouwnsmv maoouwnom Honomoe 039 use somsuon mmuoum o .m .H OH moommom :uHa moomfium> mo mamhamc4 monsoon: oouommom .mq.q maan 136 Table 4.46. Analysis of Variance (Planned Contrast) based on Citizen Action Scores between Sample Groups Administered the Solid Waste Issue Summary. Group N M SD I Env. Issues Class 81 13.827 11.174 II Intro. Bio. 38 14.394 11.068 (A - Solid Waste) III Sierra Club 10 44.900 15.242 IV Teachers (A - EE Teachers) 23 15.086 11.036 (B - Random-Sample) 28 24.642 16.598 ANOVA Groups 55 df F P Within Cells 26731.813 III vs I, II A, IV A & B 7256.344 175 47.503 .0001* I, II A vs IV A & B 1710.9003 1 11.200 .00100* I vs II A 279.228 1 1.827 .17811 IV A vs IV B 1282.663 1 8.396 .00424* * Significant relationship (p_: .05). 137 Research Question 10 Do the environmental issues class and biology subgroup II A differ significantly (p_: .05) with respect to total PECM scores and/or citizen action scores? Differences between total PECM scores and citizen action scores were determined as part of a Special Constrasts program. Results in Table 4.44 indicate no significant difference between the two groups with respect to total PECM scores. Results of citizen action scores analysis (Table 4.46) also indicated no significant difference between groups. Research Question 11 Based on total PECM scores and/or citizen action scores, do teachers as a group (groups IV A and B) differ significantly (p §_.05) from college students administered the solid waste version of the PECM? Differences between teachers and college students with respect to total PECM scores and/or citizen action scores was determined as part of a Special Contrasts program. The combined sample of college students scored significantly lower (more internal) total PECM scores than did the combined teachers group (Table 4.44). Teachers also reported taking significantly more citizen actions than the combined sample of college students (Table 4.46). Research Question 12 Do total PECM scores and/or citizen action scores for the Sierra Club differ significantly (p_: .05) from a combined sample of all 138 other subjects who responded to the solid waste version of the PECM (i.e., groups I, II A, IV A, IV B)? Results of a Planned Contrast program (Table 4.44) indicate that the Sierra Club members did not score significantly different on total PECM scores from a combined sample of subjects responding to the solid waste version of the PECM. Although differences were not significant, it should be noted that the obtained F value (F—3.1499) closely approaches significance (i.e., p=.07780) at the .05 level. An obvious limitation of this analysis was the small sample (N-lO) collected for the Sierra Club. Analysis of citizen action scores (Table 4.45) revealed that Sierra Club members reported taking significantly more actions than the combined sample of subjects. Chapter 5 DISCUSSION OF FINDINGS AND RECOMMENDATIONS This chapter first reports findings as they relate to each re- search questiont In a second section implications of findings and re- commendations for further research are presented. Discussion of Findings Research Question 1. Can evidence of reliability and validity (content and construct) be established for an instrument developed to measure perceived L of C toward taking specific categories of environ- mental action in a stated situation? The PECM is an instrument designed to measure a person's perceived L of C toward taking specific categories of environmental action in a given situation. Consistently high alpha coefficients (alpha=>.85) were achieved for total PECM scores and for each of the subscales (i.e., I, P, C — Economic, Ecomanagement, Persuasive, Political and Legal subscales). These findings indicate that the PECM exhibits strong evidence of internal consistency or reliability. Content validity of an instrument ultimately rests upon appeals to reason regarding the adequacy with which important content has been sampled and on the adequacy with which content has been cast into test items (Nunnally 1978). It is assumed the systematic procedures utilized to develOp and select PECM items substantiates evidence of the in- strument's content validity. 139 140 Initially, an item pool was developed based on objectives well— grounded in L of C and environmental action theory. Item analysis and selection were made using data collected in two pilot studies. Criteria for item retention after each pilot study was relatively high item total and respective item subscale correlations. Final PECM items did yield substantial correlations with their respective subscales and/ or with the total scale supporting the assumption of evidenced content validity. In this study construct validity is defined as the extent to which , an instrument can be shown to perform in a manner prescribed by a 1 particular construct. Construct validity cannot be claimed simply on the results of one study. An accumulation of supporting results is necessary to provide only evidence of construct validity. A number of findings in this study contribute to the support of construct validity for the PECM. These findings are discussed on the following pages and summarized following the discussion of research question 6. Generally, mean and median PECM scores for each sample group were in the lower half of the possible range of scores. This was true for total PECM scores, I, P, C subscale scores and for environmental action category subscale scores. These findings seem consistent with past L of C research. Levenson (1972a) found that very few of her subjects felt their lives were controlled by chance or powerful others to the extent that they felt they controlled their own lives. Levenson goes on to cite a number of researchers who confirm her findings (i.e., Harrow and Ferrante 1969; Hersch and Scheibe 1967; Lefcourt 1967; Rotter 1966). 141 This tendency for people to evaluate internal attributes in a more favorable light than external attributes seems to be a socially desirable response style that is inherent in L of 0 measurement (Phares 1976; Lefcourt 1976). Given that the PECM may be influenced by social desirability, the question becomes: Is the instrument contaminated to a point that its potential predictive capacity, as prescribed by the respective construct (i.e, construct validity) is seriously affected? The following dis- cussion of research questions 2-12 will provide some evidence that the PECM, regardless of the underlying effect of social desirability, is functioning in accordance with L of C/IPC theory. Research question 2. What relationship exists between scores on a measure of citizen action and total PECM scores within the environ- mental issues class, the Sierra Club group, each of the teacher: sub- groups and each of the introductory biology subgroups? Research Hypothesis: 2-1. A significant negative correlation (p_: .05) will exist between scores on a measure of citizen action and total PECM scores within each of the selected groups. Research Question 3. What relationship exists between scores on a measure of citizen action taking and total I, P, and C scores for the envirommental issues class, the Sierra Club group, each of the teacher subgroups and each introductory biology subgroup? Research Hypothesis: 3-1. A significant negative correlation (p §_.05) will exist between scores on a measure of citizen action and total I scores for each selected group. 142 3—2. A significant negative correlation (P.i .05) will exist between scores on a measure of citizen action and total P scores for each selected group. 3-3. A significant negative correlation (p-: .05) will exist between a measure of citizen action and total C scores for each selected group. Research Question 4. What relationship exists between scores on a measure of citizen action and total I+P+C scores for each environ- mental action category within the environmental issues class, the Sierra Club group, each of the teacher' subgroups, and each of the introductory biology subgroups? Research Hypothesis: 4-1. A significant negative (p j .05) correlation will exist between scores on a measure of citizen action and total I+P+C scores for each environmental action categories within each of the selected groups. Past research using relatively generalized L of C instruments has given some indication of a relationship between internality and taking environmental action (Chapter 2). Theoretically, similar findings should be realized with the PECM. Thus, it was hypothesized (preceding research questions) that total PECM scores and the various subscale scores would correlate significantly with the amount of civil actions a person reports to have taken. Civil actions are actions that en- compass procedures synomymous with.those found in the five environmental action categories. Because the total PECM and its various subscales are scored in the external direction (i.e., higher the score the more external) it was hypothesized that significant negative correlations would be achieved between the various PECM scores and citizen action scores. 143 It should also be mentioned that two different versions of the PECM were administered. All groups, with the exception of subgroup II B, received a version with a solid waste issue summary. Subgroup II B received a version with an acid rain issue summary. Pearsonjr correlations between total PECM scores and citizen action scores were consistently negative for the five groups administered the solid waste version (solid waste groups) of the PECM. In addition, a majority (four out of five) of the correlations were significant at the .05 level. Also, when data from the five solid waste groups were collapsed, analysis again produced a significant negative cor- relation between the two variables. Significant findings support the prOposed hypothesis (2-1) and indicate that, as total PECM scores move in the external direction, there was a significant drop in citizen actions reported. The one exception among solid waste groups was subgroup IV B (the random sample of teachers). Although the correlation between PECM scores and citizen action scOres was not significant (F=.23 8, p=.127), it was in the hypothesized direction and therefore provides a consistent trend for overall findings related to research question two . Correlations between total PECM scores and citizen action scores were not significant for biology subgroup II B (received the acid rain version of the PECM). However, the correlation coefficient was in the hypothesized direction (r=.2086, p=.098). Since the acid rain version of the PECM was not administered to other groups, no additional data is available to confirm or disclaim that the acid rain version is 144 functioning as hypothesized. Thus, it is difficult to determine whether results are a function of group characteristics or instrumentation. These results also raise questions as to the score of environmental issues to which the PECM can be applied. The total PECM is made up of a number of underlying subscales. Theoretically each of these subscales should show some degree of con— struct validity for it to be included in the instrument. The pattern of correlations for I, P, and C subscales was similar to that of the total PECM scores. That is, the same four out of five solid waste groups exhibited significant negative correlations between the I, P, C subscales and citizen action taking scores. Significant correlations indicate that as I, P, or C scale scores move toward the external direction (high scores), citizen actions reported decrease. Alterna— tively, as I, P, or C scores move toward the internal direction (low scores) reported citizen action taking increases. While findings for subgroup IV B were not significant, all cor- relation coefficients were in the hypothesized direction with the I and C scales falling very close to significance (respectively p=.069 and p=.078). The P scale produced a p value of .245 which was a substantial distance from the .05 significance level. Overall, correlations for all solid waste groups were in the hypothesized direction adding some consistency to findings. With respect to the acid rain group, only the C subscale scores produced a significant negative correlation with citizen action scores. This might be an indication that the C scale shows some predictive potential over the P and I scale with reference to the acid rain issue and/or this particular group. Similarly, Levenson (1972a) found that 145 only the C scale discriminated between males involved in anti-pollution activities and those not involved (i.e., those involved did not feel chance controlled their lives to the extent that those uninvolved did). Although the I and P scales did not achieve significance, the correlation coefficients were in the hypothesized direction. However, non-significant findings again raise the questions about the PECM effectiveness with different types of issues (e.g., local vs national). When the solid waste version of the PECM was broken down and analyzed, according to environmental action category subscales, evidence supporting construct validity was still present. Across the five solid waste groups each of the subscales (i.e., persuasive, political, economic, ecomanagement, legal) produced a majority (at least 3 out of 5) of significant correlations. Subscales not achieving significance differ from group to group. However, even non-significant coefficients all fall in the hypothesized direction and many approach the .05 signi- ficance level. Since each of the five subscales Show some evidence of supporting hypothesis 4-1 and thus show some evidence of construct validity, retention of each subscale within the PECM seems justified. Data analysis from the acid rain group (group II B) produced significant negative correlation coefficients for all but the legal and ecomanagement subscales. This may indicate that the other three subscales have some degree of predictive potential over the legal and ecomanagement subscales, at least with reference to the acid rain issue and/or group II B. 146 Research Question 5. What relationship exists between total I, P, and C scores within the environmental issues class, the Sierra Club group, each of the teachers subgroups and each of the introductory biology subgroups? Research Hypothesis: 5.1 A significant positive (p_: .05) correlation will exist between total I, P, and C scores within each of the groups selected (a relatively high I score refers to lack of belief in internal control). Theoretically the I, P, and C subscales all measure the same underlying construct. That is, they each measure a belief in internal or external control.- Based on this assumption and since all three subscales are scored in the negative direction, it was hypothesized that significant (p.: .05) positive correlations between the three subscales would be achieved. For all groups, I, P, and C subscales scores achieved positive and significant correlation coefficients. These findings indicate that for all groups as subjects increasingly agreed with personal control (I), their belief in powerful others and/or chance control decreased. Research Question 6. Is there a significant difference between total I, P, or C scores within the environmental issues group, the Sierra Club group, each of the teacher subgroups and each introductory biology subgroup? Research Hypothesis: 6.1 A significant difference (p i .05) in mean I, P, and C scores will be determined within sample groups. 147 The PECM was divided into I, P, and C statements based on the theory that the three subscales tap different dimensions of the same construct. If this theory were to hold true for the PECM, the diagnostic potential might be improved by interpreting individual I, P, C scores as opposed to total PECM scores. It was assumed that if significant differences among the three subscales were discovered within groups, it would indicate that the subscales are possibly measuring different L of C belief dimensions (Hypothesis 6.1). Findings indicated no significant interaction between sex and measures for any of the groups sampled. Literature on past I, P, C research neither confirms nor refutes this finding. For the environ- mental issues class, a significant F value (F=12.2379, p=.00084) indicated that females scored lower or more internal than males. Since I, P, C scores add up to make the total PECM score it is inferred that for the environmental issues group females also scored lower than males on total PECM scores. Similarly, Miller (1980; see Chapter 2 of this study), using Levenson's I, P, C scale found that within a sample of Youth Conservation Corps participants, females scored significantly lower (more internal). u s H m .H E“ E E s u A POLITICAL ACTION: An egfiomt aimed a1 pelt/sanding an g “5’ E0 E E; E0 steam/Late, a Zegialiaxm (02L KegA'AZa/the), cm execu- w m m .240 «E; 3 true goue/mmewta/K agency to confiofrm to the Vitae/s o a.) o H o H hefld by the pe/vson O/L pQ/IAOM mung that acteon, Q) <1) G) U) U) (I) a r . . , H I. H e.g., lobbying, votmg, campugmng 60ft candidates, :9 8 8° II 8 8 etc. .1: .2 :3 I: a I“. Q o a a < < l 2 3 4 5 6 l 2 3 4 5 6 1. Probably fate, more than any political action I could take, will determine the outcome of this situation. 1 2 3 4 5 6 2 I believe I can be effective in determining the outcome of this situation through the use of political action. 1 2 3 4 5 6 3. My ability to identify and utilize political actions in this situation matters little because the final outcome will be determined by a few people who will be unconcerned or unaffected by my use of such actions. 1 2 3 4 5 6 4. By participating in some type of political action, I can play an effective role in determining the outcome of this situation. 1 2 3 4 5 6 5. It would only be a coincidence if I participated in some political action related to this situation, and the situation turned out the way I felt it should. 1 2 3 4 5 6 6. If I receive training in the identification and use of political actions relevant to this situation, I could be effective in using those skills to influence the final outcome of this situation. 172 173 e. Disagree strongly hJ Disagree somewhat U) Disagree slightly #\ Agree slightly (a Agree somewhat 0‘ Agree strongly POLITICAL ACTION (continued) 10. ll. 12. 13. How this situation turns out will be determined by people above me who would be unaffected by any political action I could employ. I can implement some type of political action which would directly or indirectly influence the outcome of this situation. The political action I could take in relation to this situation would be of little or no value because it would not have an effect on the people who really decide on how this situation will turn out. If I were trained in the identification and use of political actions, I could not use those skills to influence the final outcome of this situation because it is probably uncontrollable. Even with training in the identification and use of political actions, I could not use those skills to influence the outcome of this situation because the outcome will be determined by certain people who will be unaffected by my use of such actions. I believe that what is going to happen in this situation will happen regardless of any political action I take. If I were educated in the identification and use of political actions, I could utilize that knowledge to influence the final outcome of this situation. If this situation turns out the way I believe it should, it would be the result of luck more than the result of any political action I could participate in. . 174 ,_ Disagree strongly h: Disagree somewhat U) Disagree slightly £~ Agree slightly UI Agree somewhat 0‘ Agree strongly POLITICAL ACTION (continued) N .L\ 0“ The political action I could take in this situation would be of little value in determining the outcome, because the outcome will mostly be influenced by a few people who already have . their own ideas about the situation. PERSUASIVE ACTION: An (1660/11 to umbaug mottuate hfifian Eetngo to take poottxve envtaan— mental action ao a fianctton 06 modtfited vatueb, e.g., angumentatton, debate, opeech mahtng, tettea watttng, etc. By practicing some type of persuasive action, I could play an effective role in determing the outcome of this situation. My ability to identify and utilize persuasive actions in this situation matters little because the final outcome will be determined by a few people who will be unaffected by my use of such actions. If I were experienced in the use of persuasive action techniques, it would be useless to apply those skills to this situation because the final outcome is predetermined by fate anyway. How this situation turns out will be determined by people above me who would be unaffected by any persuasive action I could take. 175 we Disagree strongly hJ Disagree somewhat to Disagree slightly $~ Agree slightly U1 Agree somewhat 0‘ Agree strongly PERSUASIVE ACTION (continued) .l.\ 10. 11. 12. 13. I believe that what is going to happen in this situation will happen regardless of any persuasive action I take. I believe I can be effective in determining the outcome of this situation through the use of persuasive action. If I were trained in the identification and use of persuasive actions, I could not use those skills to influence the final outcome of this situation because it is probably uncontrollable. Given experience in the use of persuasive action techniques, I could utilize that experience to help determine the final outcome of this situation. I believe the outcome of this situation will be influenced by what people in high social positions already think, more than by any persuasive action I could take. Fate, more than any persuasive action I could take, will determine the outcome of this situation. I can implement some type of persuasive action which would directly or indirectly in- fluence the outcome of this situation. It would be a coincidence if I participated in some persuasive action related to this situation, and the situation turned out the way I felt it should. The persuasive action I could take in this situation will be of little value in deter- mining the outcome, because the outcome will mostly be influenced by a few people who already have their own ideas about the situation. 176 rd Disagree strongly I\> Disagree somewhat b) Disagree slightly ¢~ Agree slightly UI Agree somewhat 0‘ Agree strongly PERSUASIVE ACTION (Continued) 14. 15. I believe I can acquire the ability to identify and utilize persuasive actions that would influence the outcome of this situation. Even with training in the identification and use of persuasive actions, I could not use those skills to influence the outcome of this situation because the outcome will be determined by certain people who are unaffected by my use of such actions. ECOMANAGEMENT: Any phyotcat action taken by an tnautduat on a gnoup aimed danectty at macntatnéng on tmp/Iowlng the exX/Aténg eco- oyotem, e.g., necyctL'ng, nefio/Le/statéon, e/Looton cont/Lot, commuatéve Lose 06 nuance/s, Land u/se management, etc. l 1 1 2 3 4 5 6 2 3 4 5 6 2 3 4 5 6 1. I believe I can be effective in determining the outcome of this situation through the use of ecomanagement practices. As far as I am concerned, good or bad luck will determine how this situation turns out re— gardless of my ability to identify and practice relevant ecomanagement. I can implement some type of ecomanagement strategy which would directly or indirectly influence the outcome of this situation. 177 r‘ Disagree strongly h» Disagree somewhat 0» Disagree slightly £~ Agree slightly UI Agree somewhat 0‘ Agree strongly ECOMANAGEMENT (continued) Is U1 0" 10. 11. The ecOmanagement activities I could practice in this situation would be of little value in determining the outcome, because the outcome will mostly be influenced by the actions of a few key individuals. Given experience in the identification and practice of ecomanagement techniques, I could utilize that experience to help determine the outcome of this situation. It would be a coincidence if I practiced some ecomanagement related to this situation, and the situation turned out the way I felt it should. How this situation turns out will be determined by people above me, more than by any eco- management strategies I could practice. I can have a direct or indirect effect on the quality of the environment and thus on the out- come of this situation, through the use of ecomanagement practices. The ecomanagement strategies I could practice in this situation would be of little or no value because they would not override the influence more important people than I will have on the outcome of this situation. I believe that what is going to happen in this situation will happen regardless of any ecomanagement I practice. I believe the outcome of this situation will be influenced by what people in high social positions think and do, more than by any ecomanagement strategies I could practice. 178 % .H > H o H 00 ..I: u I: 3 .I: >~ u >s O a) 00 H m H I: 8 S E "E 21" m m m :p g 8 ECOMANAGEMENT (continued) o o o H o u o o o m m m H H H w w w o o o m m m o o m m m m H H H H H H w w m o a a m < < 1 2 3 4 5 6 l 2 3 4 5 6 12. Even with training in the use of ecomanagement skills, I could not use those skills to influence the final outcome of this situation because it is probably uncontrollable. 1 2 3 4 5 6 13. Even with training in the use of ecomanagement, I could not use those skills to influence the final outcome of this situation because the outcome will be determined more by the de- cisions and actions of other people in key positions. 1 2 3 4 5 6 14. I believe I can acquire the ability to identify and practice ecomanagement strategies that would influence the Outcome of this situation. 1 2 3 4 5 6 15. If this situation turns out the way I believe it should, it would be the result of luck more than the result of any ecomanagement I could practice. ECONOMIC ACTION: Comt'etute/I an action M'm‘iam to one 05 the fiottowing: a) boycotting; b) oeiectéve coniumptton 06 goods and oe/Lviceo, e.g., puncha/se 06 necyeted mate/ziafl; c1 moneta/Ly contribution to an individuai on oaganization that activeiy wank/s 60/1 a position ouppoioted by the cont/zibutoa, e.g., donut/Loni to envi/Lonmentai came/5., membe/L- Ahip 5e2A paid to envi/Ionmen/tat activiast onganizationi, etc. 1 2 3 4 5 6 1. How this situation turns out will be determined by people above me who would be unaffected by any economic action I could take. 179 I— Disagree strongly Iv Disagree somewhat w Disagree slightly b Agree slightly UI Agree somewhat w Agree strongly ECONOMIC ACTION (continued) )—‘ N U.) D U‘I Ch 10. I believe I can be effective in determining the outcome of this situation through the use of economic actions. Even with training in the identification and use of economic actions, I could not use those skills to influence the outcome of this situation because it will be determined by people who will be unaffected by my use of such actions. It w0uld only be a coincidence if I participated in some economic actiOn related to this situation, and the situation turned out the way I felt it should. I believe the outcome of this situation will be influenced by what people in high social positions already think, more than by any economic action I could take. In this situation, I can employ some type of economic action which will have a direct or indirect effect on the final outcome. I believe that what is going to happen in this situation will happen regardless of any economic action I take. If I were to implement some type of economic action in this situation, I am sure it would have an effect on the final outcome. If I were experienced in the use of various economic actions, it would be rather useless to apply those skills to this situation because the outcome is predetermined by fate anyway. Given the experience in the identification and use of economic actions, I could utilize that experience to help determine the final outcome of this situation. 180 » u >. HIGH w .c u C3£>su>s o o m H4 m H LIE-Huge!) nova—:3: ° ‘° °° .393 3 ECONOMIC ACTION (continued) QUQ’HOU 00001030) HRH w m u: o o o o m m o o o commuuu H H -H a: m m a a a < < < 1 2 3 4 5 6 1 2 3 4 5 6 11. The economic action I could take in relation to this situation would'be of little or no value because it would not have an effect on the people who really decide how this situation will turn out. 1 2 3 4 5 6 12. As far as I am concerned, good or bad luck will determine how this situation turns out re— gardless of my ability to identify and practice relevant economic actions. 1 2 3 4 5 6 13. If I were educated in the identification and use of economic actions, I could use this knowledge to influence the final outcome of this situation. 1 2 3 4 5 6 14. The economic action I could take in this situation would be of little value in determining the outcome, because the outcome will mostly be influenced by a few people who already have their own ideas about the situation. 1 2 3 4 5 6 15. The outcome of this situation will be determined by good or bad luck, regardless of my ability to identify and utilize some type of economic action. LEGAL ACTION: Any legal/judiciary action taken by an individual and/or organization which is aimed at some aspect of environmental law enfbrcement or, a legal restrain preceding some environmental behavior perceived as undesirable, e.g., law suits, injunctions, etc. 1 2 3 4 5 6 1. If this situation turns out the way I believe it should, it would be the result of luck more than the result of any legal action I could pursue. -.- ....___ _v_. 181 r- Disagree strongly to Disagree somewhat t» Disagree slightly 9 Agree slightly Ln Agree somewhat 0‘ Agree 8 trongly LEGAL ACTION (continued) N L» .l.\ U1 0‘ The legal action I could take in this situation would be of little value in determining the outcome, because the outcome will mostly be influenced by a few peOple who already have their own ideas about the situation. If I were trained in the identification and use of legal actions, I could not use those skills to influence the final outcome of this situation becuase it is probably uncontrollable. If I receive training in the identification and application of legal actions relevant to this situation, I could be effective in using those skills to influence the final outcome of this situation. I believe the outcome of this situation will be influenced by what people in high social positions already think, more than by any legal action I could take. I believe I can be effective in determining the outcome of this situation through the use of legal actions. Even with training in the identification and application of legal actions, I could not use those skills to influence the outcome of this situation, because the outcome will be de- termined by certain peOpIe who will be unaffected by my use of such actions. I can implement some type of legal action which would directly or indirectly influence the outcome of this situation. My ability to identify and utilize legal actions in this situation matters little because the final outcome will be determined by a few peOple who will be unconcerned or unaffected by my use of such actions. 182 ,. Disagree strongly be Disagree somewhat k» Disagree slightly 4‘ Agree slightly tn Agree somewhat 0‘ Agree strongly LEGAL ACTION (continued) N b.) J.\ U1 0\ 10. 11. 12. 13. 14. 15. It would only be a coincidence if I pursued some legal action related to this situation, and the situation turned out the way I felt it should. The legal action I could pursue in this situation would have a direct or indirect effect on the final outcome. The legal action I could take in relation to this situation would be of little or no value because it would not have an effect on the people who really decide how the situation will turn out. If I were experienced in the identification and application of legal actions, it would be useless to apply those skills to this situation because the outcome is predetermined by fate anyway. By pursuing some type of legal action, I could play an effective role in determining the outcome of this situation. I believe that what is going to happen in this situation will happen regardless of any legal action I take. Continue to Section III SECTION III INSTRUCTIONS In this section you are presented with a series of questions that deal with your actual use of some problem solving actions. Please answer all the questions as completely and honestly as you can. HOW MANY TIMES DURING THE PAST TWO (2) YEARS HAVE YOU... 1... paid membership dues to an organization because you thought they would take positive action on some issue for which you have concerns. (no. of memberships) (Please list up to three (3) such organizations below) (l) (2) (3) 2... donated money (other than membership fees) to support some project related to a concern you had for an issue. _____ (no. of memberships) (Please list up to three (3) such projects below) (1) (2) (3) 3... avoided doing business with a company or refusedix>buy their product(s) in order to bring economic pressure to bear on some issue about which you have concerns. (no. of times) (Please list up to three (3) such companies or products below) (1) (2) (3) 183 184 HOW MANY TIMES IN THE LAST TWO (2) YEARS HAVE YOU... 4.. . reported to the proper authorities illegal actions taken or about to be taken by a person, group or organization. (no. of times) (Please list up to three (3) such illegal actions below) (1) (2) (3) . been involved in filing a lawsuit or filing for an injunction' concerning some issue. (no. of times) (Please list up to three (3) topics of such injunctions/lawsuits) (1) (2) (3) . participated in rallies, marches or demonstrations with the intent of persuading others to support your beliefs or actions regarding a certain issue. @o.oftmm9 (Please list below up to three (3) causes for such rallies, marches or demonstrations) (.1 ) (2) (3) 185 HOW MANY TIMES IN THE LAST TWO (2) YEARS HAVE YOU... 8... 10.. signed or distributed a petition which encourages a person, group or organization to take action on an issue about which you have concerns. (no. of times) (Please list below up to three (3) causes of such petitioning) (l) (2) (3) distributed or presented information/literature to the public about an issue for which you have concerns. (no. of times) (Please list below up to three (3) issues of such information/ literature distribution) (1) (2) (3) contacted a politician to express your support or opposition to a bill they have introduced or are considering for passage. (no. of times) (Please list up to three (3) titles or tOpics of such bills below) (1) (2) (3) How many candidates have you given of your time to campaign for over the last 5 years? (no. of candidates) 186 11... Have you used your right to vote in an attempt to improve situations (issues) about which you are concerned? Yes No Please place a check mark (pl) in front of each activity you have participated in over the last two (2) years. 1. ___ I have picked up litter and/or organized a litter campaign. 2. ___ I have taken steps to reduce energy consumption. 3. ___ I have avoided the purchase of a product because of its negative effect on the environment. 4. ___ I have taken steps to reduce my water consumption. 5. ___ I have recycled paper, glass, metals and/or organic refuse. 6. ___ I have participated in a habitat improvement project (e.g., planting shrubs for wildlife, putting up birdhouses, stream renovation). APPENDIX C Field Study (Final) PERCEIVED ENVIRONMENTAL CONTROL MEASURE (PECM) WITH BOTH ENVIRONMENTAL ISSUE SUMMARIES 187 MICHIGAN STATE UNIVERSITY DEPARTMENT OF FISHERIES AND WILDLIFE EAST LANSING ' MICHIGAN ' 48824 NATURAL RESOURCES BUILDING (517) 355-4477 Dear Participant, This is not a test. This project is intended to collect infor— mation concerning public opinions about certain aspects of social issues. The information you provide is vital to our continuing efforts to design effective environmental education programs for citizens. Your honesty in responding to the questions will be deeply appreciated. All of your responses will be kept confidential. There are three sections to this questionnaire. At the beginning of each section, you will find instructions on how to complete that section. We know that your time is very valuable and we would like to express our thanks for your cooperation on this project. Sincerely, n? (A Wipes R. Ben Peyton Assistant Professor Environmental Education Mg an... Randy Ch peau Graduate Teaching Assistant Environmental Education MSU is an Afflmatius Action/Equal Opportunity Institution SECTION I INSTRUCTIONS You are asked to read the following summary of an environ— mental issue. Please consider the situation as though it were a real part of your life and you were involved in this issue. After you have read the summary, please respond to the questions in Section II. The Issue You are a resident of the United States and have a small summer cottage on a lake in Ontario, Canada. You have just been informed by local authorities that the quality of the lake water is being threatened by increased acidity. The increased acidity is due to a phenomenon called "Acid Rain." Some research into the matter leaves you with the following information: It is generally agreed upon by researchers that "Acid Rain" is increasingly becoming a national and international problem. Acid rain is caused when sulfur oxides and nitrogen oxides —— widely recognized as among the major man—made pollutants - react with moisture in the air to form acids. Thus, when it rains, it rains a solution of water, sulfuric acid and nitric acid. Acid rain has been shown to have direct and indirect adverse effects on humans. It has been associated with decreased productivity of several natural and cultivated plant species, toxic contamination of drinking water, increased corrosion of natural and human built structures, and increased acidity of lakes, streams, and rivers. In some cases, acidity has resulted in damage to aquatic organisms and even in the complete elimination of all aquatic life. The pollutants that cause acid rain are by—products of coal, gaso- line and other fossil fuel combustion. The major sources of this combustion include industry, coal-fired power plants and the automobiles. Actually, the more fossil fuel burned the greater the amounts of pollutants discharged into the air and the more severe are the po- tential effects of acid rain. In addition, acid rain knows no boun- daries. It may fall to the ground hundreds or even thousands of miles from its initial source of production, which has caused interstate and international tension. Lines of battle are being drawn for a major conflict between those who want immediate action on acid rain and those who feel such action would block the economic growth of certain states and the nation. Legislation which calls for stricter air pollution control on indus— try, power plants and automobiles has been proposed in the national and several state legislatures. Obviously, whatever the outcome of this situation, it will have a direct or indirect effect on the quality of your life and possible on that of future generations. 188 SECTION I INSTRUCTIONS You are asked to read the following summary of an environ- mental issue. Please consider the situation as though it were a real part of your life and you were involved in this issue. After you have read the summary, please respond to the questions in Section II. The Issue You are a resident of the United States and have a small summer cottage on a lake in Ontario, Canada. You have just been informed by local authorities that the quality of the lake water is being threatened by increased acidity. The increased acidity is due to a phenomenon called "Acid Rain." Some research into the matter leaves you with the following information: It is generally agreed upon by researchers that "Acid Rain" is increasingly becoming a national and international problem. Acid rain is caused when sulfur oxides and nitrogen oxides —— widely recognized as among the major man-made pollutants - react with moisture in the air to form acids. Thus, when it rains, it rains a solution of water, sulfuric acid and nitric acid. Acid rain has been shown to have direct and indirect adverse effects on humans. It has been associated with decreased productivity of several natural and cultivated plant species, toxic contamination of drinking water, increased corrosion of natural and human built structures, and increased acidity of lakes, streams, and rivers. In some cases, acidity has resulted in damage to aquatic organisms and even in the complete elimination of all aquatic life. The pollutants that cause acid rain are by-products of coal, gaso— line and other fossil fuel combustion. The major sources of this combustion include industry, coal-fired power plants and the automobiles. Actually, the more fossil fuel burned the greater the amounts of pollutants discharged into the air and the more severe are the po— tential effects of acid rain. In addition, acid rain knows no boun— daries. It may fall to the ground hundreds or even thousands of miles from its initial source of production, which has caused interstate and international tension. Lines of battle are being drawn for a major conflict between those who want immediate action on acid rain and those who feel such action would block the economic growth of certain states and the nation. Legislation which calls for stricter air pollution control on indus- try, power plants and automobiles has been proposed in the national and several state legislatures. Obviously, whatever the outcome of this situation, it will have a direct or indirect effect on the quality of your life and possible on that of future generations. 189 SECTION II INSTRUCTIONS In this section you will find the definitions (in italics) of five approaches that might be used to solve the problem that has been described. Following each of the five definitions is a series of questions regarding your use of each approach. Read each statement and carefully circle the number at the left of each statement which best indicates how strongly you agree or disagree with the statement. Please respond to every statement. a +3 a H (d H m c u c 3 r: m +> » . , O 0 w r* m ti POLITICAL ACTION: An effort aimed at persuading L s H 42 n m . . p 0 r4 m 3 c an electorate, a legislator (or legislature), or (n m m f? E 8 executive governmental agency to conform to the Q Q’ m *4 O P values held by the person or persons taking that o o a) m w w . . . . . . L s. s action, e.g., loobying, voting, campaigning for :30 2,0 if) g g 33 candidates, etc. U) U) :0 S4 Lo L -H H m+zn m a :3 Q L: < <: < l 2 3 4 5 6 l. I believe that what is going to happen in this situation will happen regardless of any political action I take. 1 2 3 u 5 6 2. By participating in some type of political action, I can play an effective role in determining the outcome of this situation. 1 2 3 4 5 6 3. The political action I could take in this situation would be of little value in deter— mining the outcome, because the outcome will mostly be influenced by a few people who already have their own ideas about the situation. 1 2 3 4 5 6 A. I can implement some type of political action which would directly or indirectly influence the outcome of this situation. 1 2 3 u 5 6 5. If this situation turns out the way I believe it should, it would be the result of luck more than the result of any political action I could participate in. l 2 3 u 5 6 6. The political action I could take in rela- tion to this situation would be of little or no value because it would not have an effect on the people who really decide on how this situation will turn out. 190 191 a is m H (U H m n p C 3 .C >3 u h o m w H4 a H L E wa +9 Q g H 8 g;'% g o POLITICAL ACTION (Continued) r1 5 s (DCDGJI—IOJJ GJGJGJCDUJUJ s s. L m 60 w m a) m a m a a) O m m m m h S. L -H .4 H m to m D Q c3=< < < l 2 3 A 5 6 7. I believe I can be effective in determining the outcome of this situation through the use of political action. 1 2 3 4 5 6 8. How this situation turns out will be deter- mined by people above me who would be unaffected by any political action I could employ. l 2 3 A 5 6 9. It would only be a coincidence if I partici- pated in some political action related to this situation, and the situation turned out the way I felt it should. PERSUASIVE ACTION: An effort to verbally motivate human beings to take positive environmental action as a function of modified values, e.g., argumenta- tion, debate, speech making, letter writing, etc. l 2 3 A 5 6 10. Fate, more than any persuasive action I could take, will determine the outcome of this situation. 1 2 3 u 5 6 ll. I believe the outcome of this situation will be influenced by what people in high social positions already think, more than by any persuasive action I could take. I 2 3 u 5 6 l2. I believe I can be effective in determining the outcome of this situation through the use of persuasive action. l 2 3 A 5 6 13. How this situation turns out will be deter— mined by people above me who would be unaffected by any persuasive action I could take. 192 PERSUASIVE ACTION (Continued) H Disagree strongly m Disagree somewhat Ln Disagree slightly c—Agree slightly \fi Agree somewhat m Agree strongly 14. By practicing some type of persuasive action, I could play an effective role in determining the outcome of this situation. 15. The persuasive action I could take in this situation will be of little value in deter— mining the outcome, because the outcome will mostly be influenced by a few people who already have their own ideas about the situation. l6. It would be a coincidence if I participated in some persuasive action related to this situation, and the situation turned out the way I felt it should. 17. I can implement some type of persuasive action which would directly or indirectly influence the outcome of this situation. 18, I believe that what is going to happen in this situation will happen regardless of any persuasive action I take. ECOMANAGEMENT: Any physical action taken by an individual or a group aimed directly at maintain- ing or improving the existing ecosystems, e.g., recycling, reforestation, erosion control, conserva- tive use of resources, land use management, pollu- tion control, etc. 1231456 1 2 3 A 5 6 I9. I can implement some type of ecomanagement strategy which would directly or indirectly influence the outcome of this situation. 20. It would be a coincidence if I practiced some ecomanagement related to this situation, and the situation turned out the way I felt it should. 193 Disagree strongly Disagree somewhat Disagree slightly ECOMANAGEMENT (Continued) ...: N LA) t— Agree slightly \fi Agree somewhat Ox Agree strongly 21. 22. 23. 2A. 25. 26. 27. I can have a direct or indirect effect on the quality of the environment and thus on the outcome of this situation, through the use of ecomanagement praCtices. I believe that what is going to happen in this situation will happen regardless of any ecomanagement I practice. 1 I believe the outcome of this situation will be influenced by what people in high social positions think and do, more than by any ecomanagement strategies I could practice. I believe I can be effective in determining the outcome of this situation through the use of ecomanagement practices. The ecomanagement strategies I could practice in this situation would be of little or no value because they would not override the influence more important people than I will have on the outcome of this situation. If this situation turns out the way I believe it should, it would be the result of luck more than the result of any ecomanagement I could practice. The ecomanagement activities I could practice in this situation would be of little value in determining the outcome, because the out- come will mostly be influenced by the actions of a few key individuals. 194 a +> z '30 2:“ S ECONOMIC ACTION: Constitutes an action similar to g 5 "ED :3 1;} ,3: one of the following: a) boycotting; b) selective fi 5 S j; g g consumption of goods and services, e.g., purchase m w m m a) o of recycled materials; c) a monetary contribution O Q) m 2 g E to an individual or organization that actively g g: 2 m ‘0 m works for a position supported by the contributor, g a? g 8 8 8 e.g., donations to environmental causes, membership w m (n g :4 p fees paid to environmental activist organizations, -r-I w-l H £10 to ho etc Q Q C:<: < < ' l 2 3 A 5 6 28. I believe that what is going to happen in . this situation will happen regardless of any economic action I take. 1 2 3 u 5 6 29. I believe the outcome of this situation will be influenced by what people in high social positions already think, more than by any economic action I could take. 1 2 3 A 5 6 30. In this situation, I can employ some type of economic action which will have a direct or indirect effect on the final outcome. 1 2 3 A 5 6 31. If this situation turns out the way I believe it should, it would be the result of luck more than the result of any economic contri— bution I could make. l 2 3 u 5 6 32. I believe I can be effective in determining the outcome of this situation through the use of economic actions. 1 2 3 u 5 6 33. How this situation turns out will be deter— mined by people above me who would be unaffected by any economic action I could take. 1 2 3 A 5 6 34. It would only be a coincidence if I partici— pated in some economic action related to this situation, and the situation turned out the way I felt it should. 1 2 3 u 5 6 35. If I were to implement some type of economic action in this situation, I am sure it would have an effect on the final outcome. l 2 3 u 5 6 36. The economic action I could take in this situation would be of little value in deter— mining the outcome, because the outcome will mostly be influenced by a few people who already have their_own ideas about the situation. 195 Disagree strongly Disagree somewhat Disagree slightly Agree slightly Agree somewhat Agree strongly LEGAL ACTION: Any legal/judiciary action taken by an individual and/or organization which is aimed at some aspect of environmental law enforce- ment or, a legal restraint preceding some environmental behavior perceived as undesirable, e.g., law suits, injunctions, etc. N LA) J: U1 37. If this situation turns out the way I believe it should, it would be the result of luck more than the result of any legal action I could pursue. 38. I believe I can be effective in determining the outcome of this situation through the use of legal actions. 39. It would only be a coincidence if I pursued some legal action related to this situation, and the situation turned out the way I felt it should. A0. I believe the outcome of this situation will be influenced by what people in high social positions already think, more than by any legal action I could take. Al. I can implement some type of legal action which would directly or indirectly influence the outcome of this situation. A2. The legal action I could take in relation to this situation would be of little or no value because it would not have an effect on the people who really decide how the situation will turn out. A3. I believe that what is going to happen in this situation will happen regardless of any legal action I take. AA. The legal action I could pursue in this situation would have a direct or indirect effect on the final outcome. A5. The legal action I could take in this situa— tion would be of little value in determining the outcome, because the outcome will mostly be influenced by a few people who already have their own ideas about the situation. SECTION III INSTRUCTIONS In this section you are presented with a series of questions that deal with your actual use of some problem solving actions. Please answer all the questions as completely and honestly as you can. HOW MANY TIMES DURING THE PAST TWO (2) YEARS HAVE YOU... N5... A9 paid membership dues to an organization because you thought they would take positive action on some issue for which you have concerns? (no. of memberships) (Please list up to three (3) such organizations below) (1) (2) (3) donated money (other than membership fees) to support some project related to a concern you had for an issue: (no. of donations) (Please list up to three (3) such projects/donations below) (I) (2) (3) 196 197 HOW MANY TIMES IN THE LAST TWO (2) YEARS HAVE YOU... 50... 51.. 53... 5A... 55... avoided doing business with a company or refused to buy their product(s) in order to bring economic pressure to bear on some issue about which yOu have concerns. (no. of times) (Please list up to three (3) such companies or products below) (1) (2) (3) reported to the proper authorities illegal actions taken or about to be taken by a person, group or organization. (no. of times) (Please list up to three (3) such illegal actions below) (1) (2) (3) been involved in filing a lawsuit or filing for an injunction concerning some issue. (no. of times) (Please list up to three (3) topics of such injunctions/lawsuits) (l) (2) (3) 198 HOW MANY TIMES IN THE LAST TWO (2) YEARS HAVE YOU... 56... 57-- 58... 59- 60... 61... participated in rallies, marches or demonstrations with the intent of persuading others to support your beliefs or actions regarding a certain issue. (no. of times) (Please list below up to three (3) causes for such rallies, marches or demonstrations) (l) (2) (3) signed or distributed a petition which encourages a person, group or organization to take action on an issue about which you have concerns. (no. of times) (Please list below up to three (3) causes of such petitioning) (l) (2) (3) distributed or presented information/literature to the public about an issue for which you have concerns. (no. of times) (Please list below up to three (3) issues of such information/literature distribution) (1) (2) (3) 199 HOW MANY TIMES DURING THE PAST TWO (2) YEARS HAVE YOU... 6A. 65- 66. 62... contacted a politician to express your support or opposition to a bill they have introduced or are considering for passage. (no. of times) 63... (Please list up to three (3) titles or topics of such bills below) (1) (2) (3) How many candidates have you given of your time to campaign for over the last 5 years? (no. of candidates) Have you used your right to vote in an attempt to improve situations (issues) about which you are concerned? yes no Please place a check mark (J) in front of each activity you have participated in over the last two (2) years. I. I have picked up litter and/or organized a litter campaign. 2. I have taken steps to reduce energy consumption. 3. I have avoided the purchase of a product because of its negative effect on the environment. A. I have taken steps to reduce my water consumption. 5. I have recycled paper, glass, metals and/or organic refuse. 6. I have participated in a habitat improvement project (e.g., planting shrubs for wildlife, putting up birdhouses, stream renovation). , ! 67. 68. 69. 70. 200 Sex: Male Female Age: Occupation: Level of education: How long have you been a member of Sierra Club 67. 68. 69. Sex: Male Female Grade level you teach: Number of years you have taught school 67. 68. 69. Sex: Male Major in College: Year in College: Freshman Junior Grad 202 Female Sophomore Senior LIST OF REFERENCES LIST OF REFERENCES Abramowitz, S. I. "Internal-External Control and Social-Political Activism: A Test of the Dimensionality of Rotter's I-E Scale.” J. of Consulting and Clinical Psychology, 1973, 40, pp. 196-201. American Psychological AssociatiOn. Standards for Educational and Psychological Tests. Washington, D.C.: American Psycho- logical Association, 1974. Arbuthnot, J. "The Roles of Attitudinal and Personality Variables in the Prediction of Environmental Behavior and Knowledge." Environment and Behavior, 1977, g (2), pp. 217-232. Belgrade Charter. A Global Framework for Environmental Education. Developed by representatives from member nations of the United Nations during the International Environmental Education Workshop, Belgrade, Yugoslovia, October, 1975. Borden, R. J. and A. P. Schettino. "Determinants of Environmentally Responsible Behavior." J. of Environmental Education, 1979, £9 (4), pp. 35-39. Borg, W. R. and M. D. Gall. Education Research, New York: Longman Inc., 1979. Bowes, J. E.; K. R. Stamm; K. M. Jackson and J. Moore. Communications of Technical Information to Lay Audiences. Seattle, University of Washington Communication Research Center, 1978. Bowman, J. E. "Public Opinion and the Environment: Post-Earth Day Attitudes Among College Students.” Environment and Behavior, 1977,19 (3), pp. 385-416. Burrus Bammel, L.L. "Informations Effect on Attitude: A Longi— tudinal Study." J. of Environmental Education, 1978, 9 (4), pp. 41-50. Childress, R. B. "Public School Environmental Education: A National Profile." J. of Environmental Education, 1978_2 (3), pp. 2-11. 204 Council on Environmental Quality. The Global 2000 Report to the President: Volumes I and II. Superintendent of Documents. Washington, D.C. U.S. Government Printing Office, 1980. Crandall, R. B.; W. Kathkorsky; and V. J. Crandall. "Childrens Beliefs in their Control of Reinforcements in Intellectual- Academic Achievement Situations." Child Development, 1965, 36, pp. 91-109. Donovan, D. M. and M. R. O'Leary. "The Drinking-Related Locus of Control Scale: Reliability, Factor Structure and Validity." J. of Studies on Alchohol, 1978, 32 (5), pp. 759-783. Grvin, P.; G. Grvin; R. C. Lao and M. Beattie. Internal-External Control in the Motivation Dynamics of Negro Youth. J. of Social Issues, 1969, pp. 29-53. Harrow, M. and Ferrante, A. Locus of Control on Psychiatric Patients. J. of Consulting and Clinical Psychology, 1969, 33, pp. 582— 589. Hawkings and Vinton. The Environmental Classroom, Englewood Cliffs, New Jersey: Prentice Hall, 1973. Heberlein, T. A. "Social Psychological Assumptions of User Attitude Surveys: The Case of the Wilderness Scale." J. of Leisure Research, 1973, 5, pp. 18—33. Hersch, P. D. and K. E. Scheibe. "On the Reliability and Validity of Internal-External Control as a Personality Dimension." J. of Consulting Psychology, 1967, 31, pp. 609—613. Hirst, E. and L. Schuck. "An Action Program for Environmental Education." Clearinghouse, 1971, 23, pp. 203-206. Hungerford, H. and R. B. Peyton. Teaching Environmental Education. Portland, Maine: J. Weston Walch, 1976. Hungerford, H. and R. B. Peyton. "A Paradigm for Citizen Respon— sibility: Environmental Action." Current Issues VI: The Yearbook of Environmental Education and Environmental Studies. Columbus, Ohio: ERIC Center for Science, Mathematics and Environmental Education. Ohio State University, 1980, pp. 146—154. Hungerford, E.; R. B. Peyton and R. J. Wilke. "Goals for Curriculum Development in Environmental Education." J. of Environmental Education, 1980, Il_(3), pp. 42-47. Kleiber, D.; D. J. Veldman; and S. L. Menaker. "The Multidimen— sionality of Locus of Control." Paper presented at Eastern Psychological Association Convention, Washington, D.C., 1973. Koenig, D. J. "Additional Research on Environmental Activism." Environment and Behavior, 1975, Z_(4), pp. 472—485. Lefcourt, H. Belief in Personal Control: Research and Implications. Journal of Individual Psychology, 1966, 22, pp. 185-195. Lefcourt, L. M. Locus of Control: Current Trends in Theory and and Research. Lawrence Earlbaum Associates, Inc., 1976 Loret, J. A. Practical Model for EnvirOnmental Education. J. of Environmental Education, 1974, 5, pp. 33-36. Levenson, H. Locus of Control and Other Cognative Correlates of Involvement in Anti—Pollution Activities. Unpublished Doctoral Dissertation. Claremont Graduate School, Claremont, CA, 1972(a). Levenson, H. Distinctions within the Concept of Internal—External Control: Development of a New Scale. Proceedings, 80th Annual Convention, APA, 1972, 7, pp. 261—262. (b) Levenson, H. "Perceived Parental Antecedents of Internal, Powerful Others and Chance Locus of Control Orientations." Develop— mental Psychology, 1973, 9, pp. 268—274. (a) Levenson, H. Reliability and Validity of the I, P, and C Scales — A Multidimensional View of Locus of Control. Paper presented at American Psychological Association Convention, Montreal, August, 1973. Pub. ERIC, Center for Science, Mathematics and Environmental Education, 1973, ED 087—791. (b) Levenson, H. "Multidimensional Locus of Control in Psychiatric Patients.” J. of Consulting and Clinical Psychology, 1973, 41, pp. 397—404. (c) Levenson, H. "Activism and Powerful Others: Distinctions within the Concept of Internal—External Control." J. of Personality Assessment, 1974, 38, pp. 377-383. Mason, P. F. "Theory and Practice of Environmental Impact Analysis," J. of Environmental Education, 1974, 6 (2), pp. 40—44. Nunnally, J. C. Psychometric Theory. McGraw—Hill Inc., 1978. Opperheim, A. N. uestionnaire Desi and Attitude Measurement, New York: Basic Books, Inc. 1966. Peyton, R. B. An Assessment of Teachers' Abilities to Identify, Teach and Implement Environmental Action Skills. Unpublished Doctoral Dissertation. Dept. of Curriculum, Instruction, and Media, Southern Illinois University, Carbondale, IL, 1977. 206 Peyton, R. B. and B. A. Miller. "Developing an Internal Locus of Control as a Prerequisite to Environmental Action Taking.” Current Issues VI: The Yearbook of Environmental Education and Environ- mental Studies. Columbus, Ohio: ERIC Center for Science, Mathematics and Environmental Education. Ohio State University, 1980, pp. 173-192. Phares, E. J. Locus of Control in Personality, Morristown: General Learning Press, 1976. Ramsey, C. E. and R. E. Rickson. "Environmental Knowledge and At— titudes." J. of Environmental Education, 1977, §_(1), pp. 10—18. Reid, D. W. and E. E. Ware. "Multidimensionality of Internal Versus External Control: Addition of a Third Dimension and Non— distinctions of Self Versus Others." Canad. J. Behavior Science, 1974 (6, (2), pp. 131—142. Rilo, T. J. "Basic Guidelines for Environmental Education." J. of Environmental Education, 1974, 6, pp. 52—55. Rotter, J. B. Social Learning Theory and Clinical Psyghology. Englewood Cliffs, NJ; Prentice Hall, 1954. . "Generalized expectancies for internal versus external control of reinforcement." Psychological Monographs, 1966, 80. Rotter, J. B.; Chance, J., and Phares, E. J. (Eds.) Applications of a Social Learning Theory of Personality. New York; Holt, Rinehart and Winston, 1972. Rotter, J. B. "Some Problems and Misconceptions Related to the Con— struct of Internal Versus External Control of Reinforcement." J. of Consulting and Clinical Psychology, 1975, 43 (1), pp. 56-67. Sarger, S. P. and H. A. Alker. "Dimensions of Internal-External Locus of Control and the Womens Liberation Movement." J. of Social Issues. 1972, 28, pp. 115—129. Schoenfeld, C. A. "Earth Day '70 '80 '90" Current Issues VI: The Yearbook of Environmental Education and Environmental Studies. Columbus, Ohio: ERIC Center for Science, Mathematics and Environmental Education. Ohio State University, 1977, pp. xvii-xxvi. Secman, M. and J. Evans. "Alienation and Learning in a Hospital Setting.’ American Sociological Review, 1962 27, pp. 772-782. 207 Sharma, N. C.; Kivlin J. E. and Fliegel, F. C. "Environmental Pollution: Is there Enough Public Concern to Lead to Action?" Environment and Behavior, 1975, Z_(4), pp. 455-471. Smith, J. M. A Study of the Variables Associated with Environmental Action and Locus of Control Among Four Dissimilar Populations. Unpublished Masters Thesis. Southern Illinois University, Carbondale, IL., 1979. Stamm, K. R. and J. E. Grunig. "Communication Strategies and Cognative Strategies in Resolving Environmental Issues." Journalism Quarterly, 1977 (Winter). lpp. 713-728. Stapp, W. B. "Environmental Encounters." In Clay Schoeflt (ed.) Outlines of Environmental Education, Madison, WI; Dembar Educational Research Services, Inc., 1971. Stapp, W. B. and D. A. Cox. ”Environmental Education Activities Manual." Farmington Hills, Michigan: William B. Stapp and Dorthy A. Cox, 1975. Tbilisi Intergovernmental Conference on Environmental Education. Toward an Action Plan: A Report on the Tbilisi Conference on Environmental Education. A paper developed by the FICE sub- committee on Environmental Education. Washington, D.C. U.S. Government Printing Office, stock 017—080—01838—1, 1978. Tittle, C. R. and Hill, R. J. ”Attitude Measurement and Prediction of Behavior: An Evaluation of Conditions and Measurement Techniques." Sociometry, 1976, 30, pp. 199-213. Tomera, A. N. Environmental Locus of Control Instrument. Unpublished institutional research, Department of Curriculum, Instruction and Media, Southern Illinois University, Carbondale,IL, 1979. Tuckman, B. W. Conducting Educational Research. Harcourt Brace Jovanoick, Inc. 1978. Tucker, L. R. "The Environmentally Concerned Citizen Some Correlates." Environment and Behavior, 1978, 29 (3). pp. 389—418. Walkey, F. H. "Internal Control, Powerful Others, and Chance: A Conformation of Levenson's Factor Structure." J. of Personality Assessment, l979,_£3, (5), pp. 532-535. Wallston, B. S.; Walston, K. A.; Kaplun G. D.; Maides, S. A. Develop- ment and Validation of the Health Locus of Control Scales." J. of Consulting and Clinical Psychology, 1976, 44 (4), pp. 580—585. IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIZIIIIIIIIIIIIIIIIIIIIIIII