THE RELATIONSHIP BETWEEN SELF-EFFICACY AND AUTONOMOUS LEARNING Michael K. Ponton, M. Gail Derrick, Paul B.

Carr, and James Michael Hall

Abstract Previous work has shown that cognized goals and outcome expectancies motivate engagement in specific behaviors through the mediating influence of self-efficacy. Autonomous learning represents a specific set of behaviors as measured by intentions to show resourcefulness, initiative, and persistence in ones learning; thus, self-efficacy in autonomous learning should precede a learners participation in autonomous learning activities. The purpose of the present study was to develop a self-efficacy in autonomous learning instrument thereby enabling future research to test this hypothesized causal relationship. Two pilot studies were conducted in this developmental process: the first pilot study (N = 77) focused on instrument parsimony while the second study (N = 51) was conducted to confirm item homogeneity and internal consistency of the resultant instrument. The results suggest that the final form of the Appraisal of Learner Autonomy is both valid and reliable thereby enabling future research into this psychological factor associated with learner autonomy.

Introduction In 1992, Confessore asserted that in order for self-directed learning to lead to a personally satisfying conclusion, the factors of desire, resourcefulness, initiative, and persistence must be manifest. Over the past 5 years, researchers have developed valid instrumentation to assess these factors (Carr, 1999; Derrick, 2001; Meyer, 2001; Ponton, 1999), with continuing levels of refinement (cf. Park & Confessore, 2002), in an attempt to supply the field of self-directed learning with new research-based insights into these aspects of autonomous learning. Ponton (1999) defined learner autonomy as the characteristic of the person who independently exhibits agency [i.e., intentional behavior] in learning activities (pp. 1314) and stated that autonomy represents a subset of the attributes associated with selfdirectedness. Ponton suggested that autonomy, like self-directedness, represent cognitive and affective qualities of the agent while autonomous learning refers to subsequent conative manifestations. The term conative is used with aspects of autonomous learning because conation refers to his [sic, i.e., the agents] behavioral intentions (Fishbein &Ajzen, 1975, p. 12) and ensuing intentional activity. A recent path-analytic study conducted by Ponton, Carr, and Derrick (2003), however, concluded that a measure of self-efficacy in autonomous learning was needed to better describe the relationship between motivation and conation (p. 13). This recommendation was based upon the lack of context with respect to adult autonomous learning associated with Meyers (2001) desire measure and the important role of selfefficacy as a predictor of human performance (Bandura & Locke, 2003). Meyer created

2 the Inventory of Learner Desire (ILD) as an assessment of the degree to which an agent can act intentionally, independent of any particular contextual manifestation; thus, its explanatory utility within the context of autonomous learning was questioned (Ponton, Carr, & Derrick, 2003). Contrary to the ILD, the Inventory of Learner Resourcefulness (ILR; Carr, 1999), the Inventory of Learner Initiative (ILI; Ponton, 1999), and the Inventory of Learner Persistence (ILP; Derrick, 2001) were constructed to assess intentions of respective subscales within the domain of adult learning. Hoban and Sersland (1998) performed a study to determine if a correlation existed between readiness for self-directed learning (assessed via the Self-Directed Learning Readiness Survey) and self-efficacy for self-directed learning. As part of this investigation, the 10-item Self-Efficacy for Self-Directed Learning Questionnaire (SSLQ) was developed. A statistically significant correlation was found (r = 0.49, N = 86, p < .001) between the two measures for a sample of students who were tested at the beginning of their teacher credentialing program. This research suggested that selfefficacy is an important construct in understanding a students readiness for self-directed learning. For the purposes of investigations on the larger population of adult learners, however, the SSLQ is inadequate due to items that are expressly related to structured education. References to teacher (item 2, Hoben & Sersland, 1998, p. 17), fellow students (item 4, p. 17), student directed cooperative groups (item 6, p. 17), instructional videotapes (item 8, p. 18), and graduate courses (item 10, p. 18) suggest that the SSLQs validity is related to its use on adults participating in formal graduate education. However, the concept of autonomous learning as an agentive activity is not limited to students participating in educational programs. While such students certainly can exhibit autonomy in their learning, other adults may do so as well. Thus, a new instrument was required to measure self-efficacy within the construct of autonomous learning. The purpose of the present study was to develop an instrument to measure selfefficacy in autonomous learning. Such an instrument would use the guidelines presented by Bandura (2001) in developing self-efficacy scales and testing would be performed to support validity and reliability. The hypothesized role of self-efficacy with autonomous learning will be presented along with results from the instrument development activity.

Desire, Self-Efficacy, and Autonomous Learning Fishbein and Ajzen (1975) provided a simple model relating beliefs, attitudes, intentions, and behaviors. They described beliefs as a cognitive process in which objects are assigned attributes that provide a discrimination function; that is, cognition is ones knowledge. Attitudes are a learned affection of favor or disfavor with respect to different objects based upon the discriminating attributes where such attitudes may influence (in addition to being influenced by) beliefs. Attitudes toward objects then influence the intentions of the agent and subsequent behaviors where the consequences of behaviors provide feedback to ones belief system. This model is presented in Figure 1.

Beliefs (Cognition)

Attitudes (Affection)

Intentions (Conation)

Behaviors

Figure 1. The model of Fishbein and Ajzen (1975).

Based upon past research, Bandura (1997) summarizes the mediating influence of self-efficacy on cognitive motivation. An agent will not engage in performances and adopt performance goals that are thought to lead to desirable outcomes unless the agent feels efficacious in effecting a successful performance (cf. Ponton, Edmister, Ukeiley, & Seiner, 2001). People do not engage in perceived futile endeavors but rather choose activities that they feel will lead to satisfying ends. Through self-reflection, people process different sources of efficacy information (i.e., mastery experiences, vicarious experiences, verbal persuasion, and physiological/emotive arousals) and form beliefs about their level of capability in diverse arenas. Using this knowledge, favorable or unfavorable attitudes toward engagement in select behaviors are developed that influence activity choice. Thus, self-efficacy precedes conation. Self-efficacy is a domain specific assessment. Ones perception of capability in turning on a light switch is typically much different than an assessment of capability in running a 4-minute mile. Therefore, assessments of efficacy must be contextualized to the activity of interest. The context of autonomous learning is no different. If one believes that engagement in autonomous learning activities will lead to desirable outcomes and feels capable of successfully doing such learning, then it should be expected that selfefficacy in autonomous learning should precede such learning. As previously mentioned Meyers (2001) instrument of desire is not contextualized to adult autonomous learning but rather is an assessment of an adults ability to exert influence over his or her life by considering the three subscales of freedom, power, and change. Her instrument represents an attempt to measure the degree to which an agent can act intentionally (Ponton, Carr, & Derrick, 2003, p. 2) or as Park and Confessore (2002) asserted, [Meyers] work on desire to learn has been treated as an effort to understand the precursors to the development of intentions related to learning (p. 289). Unfortunately, without a behavioral context, ones general belief concerning the ability to act intentionally will probably provide little explanatory utility in understanding specific conative manifestations such as autonomous learning. If autonomous learning is determined by a summation of the ILR, ILI, and ILP and using the presented arguments of context, it is hypothesized that the largest effect on autonomous learning will be through the path Desire Self-Efficacy in Autonomous Learning Autonomous Learning

4 where self-efficacy should mediate the influence of desire on autonomous learning. Because of the past lack of suitable self-efficacy instrumentation, research on testing the hypothesized path analytic model represents future work.

Development of the Appraisal of Learner Autonomy First Pilot Study The Appraisal of Learner Autonomy (ALA; Appendix A) was constructed using the ideas presented by Bandura (2001) and modeled after the Exercise Self-Efficacy Scale (ESS) (Bandura, 2001, p. 11) by permission (Bandura, personal communication, October 6, 2003). As per Banduras (2001) guidelines, the title of the ALA does not include selfefficacy to avoid response bias (Bandura, p. 6), the scale is in gradations of can do to reflect a measure of capability (Bandura, p. 4), the scale ranges from 0-100 to improve predictive utility (Pajares, Hartley, & Valiante, 2001), the text is domain specific to autonomous learning to improve both predictive and explanatory utility within this construct (Bandura, p. 1), and the items represent performance impediments to maximize discrimination between respondents (Bandura, p. 3). The introductory statement in the ALA reads: please rate how sure you are that you can get yourself to participate in a learning activity when nobody else requires you to do so. Note that a learning activity is any [sic] activity that you believe will help you to learn something that you want to learn. (Appendix A, 1) This statement reflects the position of Ponton, Carr, and Confessore (2000) that autonomous learning is an agentive activity where agency refers to behavior that is intentional and based upon a multitude of sociocognitive determinants. Self-efficacy is asserted to be relevant to autonomous learning as manifest agency because perceived self-efficacy is an important part of that constellation of unmeasured determinants of performance (Bandura & Locke, 2003, p. 91). Social cognitive theory posits the existence of three types of barriers to agency: cognitive (i.e., self-inefficacy), situational (i.e., temporary), and structural (i.e., inadequate resources) (Bandura, 1997); however, in general, the items presented in the ESS scale include situational impediments applicable to any type of adult activity. The original 21-item version of the ALA (Appendix A) included 20 items directly associated with the complete ESS that were either (a) taken verbatim (ALA items 1-3, 7, 8, 11-15, 19, and 21), (b) contextualized to learning as opposed to exercise (ALA items 4, 9, and 10), (c) separated to avoid double-barreled items (ALA items 5, 6, 17, 18), or (d) rewritten slightly (ALA item 16). Only ALA item 20 is new and reflects a structural barrier (i.e., monetary cost) as suggested by Darkenwald and Valentine (1985) as a deterrent to adult learning. Bandura (2001) argues that self-efficacy scales are face valid; however, research directed at establishing predictive validity (e.g., future work with self-efficacy and autonomous learning) help support a scales validity. Bandura suggests the use of factor analysis to determine item homogeneity (i.e., content validity) and Cronbachs alpha coefficient as a measure of internal consistency (i.e., reliability). A pilot test was

5 performed on the original 21 items to determine a resultant instrument for the present study. The first pilot group consisted of 77 participants selected (i.e., several convenience samples) from Regent University (n = 51) and Arkansas State University at Jonesboro (n = 26). With respect to the Regent University participants, some participants were staff members and not all student participants were from a given academic discipline (approximately 60% of the participants were students studying in the library and thus unknown to the researchers). All Arkansas State University participants were inservice teachers. The demographics of this pilot group are presented in Table 1 where the majority of the participants were female (P = 65%), White (P = 84%), and had a bachelors degree (P = 56%). The average age of this pilot group was 34.96 years (SD = 12.13) and ranged from 21 to 63.

Table 1 Pilot Group 1 (N = 77) Demographics Variable Gender Male Female Racial/Ethnic Group White Black Hispanic Other Highest Educational Attainment High School Diploma Bachelors Degree Graduate Degree n 27 50 65 7 2 3 14 43 20

Descriptive statistics for each of the original 21 items is presented in Table 2. Because no item means are 100 (i.e., certain can do), it can be asserted that the ESS impediments represent obstacles within the context of adult autonomous learning. Because homogeneity of items is hypothesized to exist, principal component analysis was performed to uncover uncorrelated common factors that account for a maximum amount of variance in the subgroup of items (Dunteman, 1989) where the first principal component represents the best condensation of the variables (Gorsuch, 1983). However, because the principal component analysis was performed on the correlation matrix and because parsimony was desired, items of non-normal character were identified for removal using the Kolmogorov-Smirnov Z (K-S Z) goodness-of-fit test. Because the score for each participant will be the summation of all items, the negative implications of committing a Type I error for each items respective K-S Z test were assumed to be minimal; therefore, the K-S Z test of normality was based on = 0.1. The results (see Table 2) suggest that items 1, 2, 4, 5, 7, 14, 15, 16, and 19 may be assumed to follow a normal distribution. The resultant 9-item ALA is presented in Appendix B.

6 Table 2 Descriptive Statistics and Kolmogorov-Smirnov Z Test of the Original 21 Items for Pilot Group 1 Item 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 M 62.16 66.14 81.94 66.56 58.56 74.69 58.51 65.55 72.47 57.78 57.22 78.52 67.04 57.45 65.00 68.05 72.99 80.78 64.35 77.99 68.68 SD 24.01 24.96 18.71 24.55 25.64 20.50 26.77 25.55 21.88 26.86 32.43 22.95 22.65 28.68 24.51 23.97 25.73 21.21 24.98 22.28 23.67 Skewness -0.36 -0.50 -1.12 -0.54 -0.38 -0.95 -0.36 -0.81 -0.80 -0.47 -0.33 -1.23 -0.76 -0.53 -0.77 -0.44 -1.03 -1.50 -0.47 -1.37 -0.72 Kurtosis -0.43 -0.62 0.71 -0.45 -0.44 0.68 -0.61 -0.04 0.20 -0.84 -1.17 0.87 0.06 -0.68 0.08 -0.61 0.45 2.38 -0.46 1.86 -0.21 K-S Z 0.997 0.994 1.748 1.017 0.885 1.638 1.173 1.462 1.258 1.483 1.288 1.738 1.500 1.165 1.222 1.163 1.569 1.646 1.203 1.626 1.422 p 0.274 0.277 0.004* 0.252 0.413 0.009* 0.127 0.028* 0.084* 0.025* 0.073* 0.005* 0.022* 0.132 0.101 0.134 0.015* 0.009* 0.111 0.010* 0.031*

Note. Skewness SE = 0.27; kurtosis SE = 0.54. K-S Z refers to the Kolmogorov-Smirnov Z test with associated p-values. *p < 0.1.

Because principal component analysis was performed on the correlation matrix, results may be dependent upon sampling variability (Kim & Mueller, 1978). According to Kline (1993), a minimum subject-to-item ratio necessary for good factor analysis is 2:1 (p. 121) although he states that there are large disagreements among researchers concerning this ratio. However, because the pilot group sample size to item number was approximately 9:1 (i.e., 77:9), the sample size was assumed to be sufficient for principal component analysis. Table 3 presents the results of the principal component analysis performed on the nine items where normality was assumed. (Note that the item numbers in Table 3 correspond to the 21-item instrument presented in Appendix A and not the reduced 9item instrument presented in Appendix B.) The Bartlett Test of Sphericity and the KaiserMeyer-Olkin Measure of Sampling Adequacy (MSA) were used to assess the suitability of the correlation matrix for factor analysis. For factor analysis, the MSA index should be greater than 0.5 (Cureton & DAgostino, 1983, p. 389) and Bartletts 2 should have a low p-value thereby enabling a rejection of the null hypothesis of no difference between the correlation matrix and the identity matrix (Norusis, 1988). As presented in Table 3, MSA = 0.84 and Bartletts 2(36, N = 77) = 322.98 with p < 0.001; thus, the pilot group

7 sample was assumed adequate for principal component analysis. Gorsuch (1983) indicates that a minimum factor loading level of 0.3 (p. 210) is popularly used to define a salient loading. The results presented in Table 3 indicate that all nine items load above this criterion (minimum loading is 0.546) thereby supporting item homogeneity.

Table 3 Principal Component Analysis of the Reduced 9-Item Instrument for Pilot Group 1 Item 1 2 4 5 7 14 15 16 19 Factor Loading 0.817 0.761 0.546 0.810 0.814 0.644 0.713 0.633 0.728

Note. Only one component extracted explaining 52.43% of the total variance. Kaiser-Meyer-Olkin Measure of Sampling Adequacy (MSA) = 0.84; Bartletts Test of Sphericity approximate 2(36, N = 77) = 322.98, p < 0.001.

As suggested by Bandura (2001), Cronbachs alpha coefficient was computed to assess internal consistency. Kline (1993) suggests that alpha should be high at around 0.9 but not less than 0.7 (p. 11). Using the pilot group data, Cronbachs alpha was computed to equal 0.88 for the nine items associated with Table 3. Based upon an analysis of the results of the pilot study, the ALA was reduced to nine items. These items were found to be homogeneous via principal component analysis and internally consistent via Cronbachs alpha coefficient thereby supporting content validity and internal reliability, respectively. The resultant items are presented in Appendix B and were used for a second pilot study. Second Pilot Study A second pilot study was conducted using the 9-item version of the ALA. Fifty one graduate students attending Regent University participated in the second study. The purpose of this second study was to determine if the observed item homogeneity and internal consistency found in the first pilot study was present when the shortened ALA was used. The 9-item ALA (Appendix B) was sent to all graduate students (approximately 500) attending programs in Regent Universitys School of Education (see Table 4). The majority of the second pilot group was female (P = 80%), White (P = 69%), and had a graduate degree (P = 69%).

8 Table 4 Pilot Group 2 (N = 51) Demographics Variable Gender Male Female Racial/Ethnic Group White Black Hispanic Other Highest Educational Attainment High School Diploma Bachelors Degree Graduate Degree n 10 41 35 13 1 2 0 16 35

The descriptive statistics for the second pilot study are presented in Table 5. Note that item numbers 1 through 9 (Appendix B) correspond to item numbers 1, 2, 4, 5, 7, 14, 15, 16, and 19 (Appendix A), respectively. Comparing Table 5 to Table 2, the means for all nine items for pilot group 2 was larger than for pilot group 1. Because parsimony was not a goal of the second pilot study, item reduction via normality analysis was not performed.

Table 5 Descriptive Statistics for the 9-Item ALA from Pilot Group 2 Item 1 2 3 4 5 6 7 8 9 M 70.35 76.27 72.37 72.67 67.74 65.39 79.31 75.96 71.18 SD 21.76 18.70 19.83 20.21 21.66 24.37 18.52 20.79 22.13

The results of the principal component analysis are presented in Table 6. For this analysis, MSA = 0.76 and Bartletts 2(36, N = 51) = 217.38 with p < 0.001; thus, the second pilot group sample was assumed adequate for principal component analysis (Cureton & DAgostino, 1983, p. 389; Norusis, 1988). According to the 0.3 criterion of Gorsuch (1983), all nine items have salient factor loadings; thus, items homogeneity is

9 tenable. Cronbachs alpha coefficient was computed to equal 0.86 for the second pilot group; thus, internal consistency is supported as well.

Table 6 Principal Component Analysis of the 9-Item ALA for Pilot Group 2 Item 1 2 3 4 5 6 7 8 9 Factor Loading 0.707 0.734 0.656 0.797 0.775 0.623 0.765 0.490 0.718

Note. First component explaining 49.28% of the total variance. Kaiser-Meyer-Olkin Measure of Sampling Adequacy (MSA) = 0.76; Bartletts Test of Sphericity approximate 2(36, N = 51) = 217.38, p < 0.001.

Conclusion Based upon two pilot studies, the 9-item Appraisal of Learner Autonomy appears to be a valid and internally reliable instrument. Further research can now be undertaken in conjunction with the Learner Autonomy Profile to determine the tenability of the hypothesized causal relationship that self-efficacy mediates the influence of desire on the conative manifestations of resourcefulness, initiative, and persistence in autonomous learning. Support for this model will provide a greater level of understanding into methods of fostering learner autonomy via the sources of efficacy information. The present investigation was performed as a logical next step in ongoing research to understand the psychological aspects of autonomous learning. Without adequate instrumentation, conjectures will not lead to tenable theories that support future empiricism into uncovering viable methods of empowering agents to further achievement. Self-efficacy has been argued to mediate all forms of cognitive motivation; thus, a research-based understanding as to its role in autonomous learning is essential if we are to continue to move forward in developing lifelong learners.

Acknowledgements The authors would like to thank Professor Albert Bandura (Stanford University) for his permission in using the Exercise Self-Efficacy Scale as the model for the Appraisal of Learner Autonomy Scale.

10 References Bandura, A. (1997). Self-efficacy: The exercise of control. New York: W. H. Freeman and Company. Bandura, A. (2001). Guide for constructing self-efficacy scales. Available from Albert Bandura, Department of Psychology, Stanford University, Standford, CA 943052130. Bandura, A., & Locke, E. A. (2003). Negative self-efficacy and goal effects revisted. Journal of Applied Psychology, 88(1), 87-99. Carr, P. B. (1999). The measurement of resourcefulness intentions in the adult autonomous learner (Doctoral dissertation, The George Washington University, 1999). Dissertation Abstracts International, 60, 3849. Confessore, G. J. (1992). An introduction to the study of self-directed learning. In G. J. Confessore & S. J. Confessore (Eds.), Guideposts to self-directed learning: Expert commentary on essential concepts (pp. 1-6). King of Prussia, PA: Organization Design and Development, Inc. Cureton, E. E., & DAgostino, R. B. (1983). Factor analysis: An applied approach. Hillsdale, NJ: Lawrence Erlbaum Associates. Darkenwald, G. G., & Valentine, T. (1985). Factor structure of deterrents to public participation in adult education. Adult Education Quarterly, 35(4), 177-193. Derrick, M. G. (2001). The measurement of an adults intention to exhibit persistence in autonomous learning (Doctoral dissertation, The George Washington University, 2001). Dissertation Abstracts International, 62, 2533. Dunteman, G. H. (1989). Principal components analysis. Newbury Park, CA: Sage Publications. Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavior: An introduction to theory and research. Reading, MA: Addison-Wesley. Gorsuch, R. L. (1983). Factor analysis. Hillsdale, NJ: Lawrence Erlbaum Associates. Hoban, G. J., & Sersland, C. J. (1998, February). Developing learning plans for adult learners: Can self-efficacy predict a rediness for self-directed learning to determine effective modes of instruction? Paper presented at the 12TH International Symposium on Self-Directed Learning, Kissimmee, FL. Kim, J., & Mueller, C. W. (1978). Introduction to factor analysis: What it is and how to do it. Beverly Hills, CA: Sage Publications. Kline, P. (1993). The handbook of psychological testing. New York: Routledge. Meyer, D. A. (2001). The measurement of intentional behavior as a prerequisite to autonomous learning (Doctoral dissertation, The George Washington University, 2001). Dissertation Abstracts International, 61, 4697. Norusis, M. J. (1988). SPSS-X advanced statistics guide (2nd ed.). Chicago: SPSS Inc. Pajares, F., Hartley, J., & Valiante, G. (2001). Response format in writing self-efficacy assessment: Greater discrimination increases prediction. Measurement & Evaluation in Counseling & Development, 33(4). Retrieved January 7, 2004, from http://www.emory.edu/EDUCATION/mfp/PHV2001MECD.html Park, E., & Confessore, G. J. (2002). Development of new instrumentation: Validation of the Learner Autonomy Profile Beta version. In H. B. Long & Associates (Eds.),

11 Twenty-first century advances in self-directed learning (pp. 289-306). Schaumburg, IL: Motorola University Press. Ponton, M. K. (1999). The measurement of an adults intention to exhibit personal initiative in autonomous learning (Doctoral dissertation, The George Washington University, 1999). Dissertation Abstracts International, 60, 3933. Ponton, M. K., Carr, P. B., & Confessore, G. J. (2000). Learning conation: A psychological perspective of personal initiative and resourcefulness. In H. B. Long & Associates (Eds.), Practice & theory in self-directed learning (pp. 65-82). Schaumburg, IL: Motorola University Press. Ponton, M. K., Carr, P. B., & Derrick, M. G. (2003, February). A path analysis of the conative factors associated with autonomous learning. Paper presented at the 17th International Self-Directed Learning Symposium, Cocoa Beach, FL. Ponton, M. K., Edmister, J. H., Ukeiley, L. S., & Seiner, J. M. (2001). Understanding the role of self-efficacy in engineering education. Journal of Engineering Education, 90(2), 247-251.

12 APPENDIX A
Appraisal of Learner Autonomy [Pilot Group Version] In responding to the items below, insert any score (0-100) using the following scale: 0 Cannot do at all 10 20 30 40 50 60 Moderately certain can do 70 80 90 100 Certain can do

In each of the following situations, please rate how sure you are that you can get yourself to participate in a learning activity when nobody else requires you to do so. Note that a learning activity is any activity that you believe will help you to learn something that you want to learn. (0-100) 1. When I am feeling tired 2. When I am feeling under pressure from work 3. During bad weather 4. After recovering from an injury that interrupted my learning 5. When I am experiencing personal problems 6. After I have experienced personal problems 7. When I am feeling depressed 8. When I am feeling anxious 9. After recovering from an illness that interrupted my learning 10. When I feel physical discomfort during my learning activity 11. During a vacation 12. After a vacation 13. When I have too much work to do at home 14. When visitors are present 15. When there are other interesting things to do 16. When I am not getting near my learning goals 17. Without support from my family 18. Without support from my friends 19. When I have other time commitments 20. When I have a limited amount of money 21. After experiencing family problems _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____

13 APPENDIX B
Appraisal of Learner Autonomy [Final Version] In responding to the items below, insert any score (0-100) using the following scale: 0 Cannot do at all 10 20 30 40 50 60 Moderately certain can do 70 80 90 100 Certain can do

In each of the following situations, please rate how sure you are that you can get yourself to participate in a learning activity when nobody else requires you to do so. Note that a learning activity is any activity that you believe will help you to learn something that you want to learn. (0-100) 1. When I am feeling tired 2. When I am feeling under pressure from work 3. After recovering from an injury that interrupted my learning 4. When I am experiencing personal problems 5. When I am feeling depressed 6. When visitors are present 7. When there are other interesting things to do 8. When I am not getting near my learning goals 9. When I have other time commitments _____ _____ _____ _____ _____ _____ _____ _____ _____