Reactions to the 2008 economic crisis and the theory of planned behavior.

Chambers, Valrie ; Benibo, Bilaye R. ; Spencer, Marilyn 等

INTRODUCTION

By the early fall of 2008, all mainstream US news media began warning that problems experienced in financial institutions were having a detrimental effect on Wall Street and were threatening the stability of at least some banks. They reported on high level, urgent meetings of the Secretary of the Treasury and the Chairman of the Federal Reserve System with the heads of federal agencies and investment banks. In that environment, many middle income Americans saw the value of their financial portfolios decrease significantly, and others feared that their savings were in jeopardy. All of this psychic pain provided a unique quasi-experiment for attempts to learn about the effects of perceptions on investing and saving behavior.

Peoples' intentions and actions, in aggregate can shift economic markets, and not always in a good way. A deeper analysis is needed to understand what factors influence intentions and actions. The theory of planned behavior asserts that people think first (intend) and then act. This theory has been successfully applied to predicting actions in a wide variety of decisions and outcomes, including losing weight (Ajzen, 1991) and computer resource center usage by business students (Taylor and Todd, 1995). In the theory of planned behavior, attitudes, perceived behavioral control, self-efficacy, and behavioral norms are all dependent variables of intent to act, which in turn is a dependent variable to actual behavior. In this paper, we examine its usefulness for predicting how people intend to react (with respect to their employment and investment strategies) to a perceived national economic crisis. In a meta-study of the link between intent and action, Sheppard, et al. (1988) found the link between these two variables to be both significant and robust in size. The rest of the paper is organized as follows: relevant literature concerning the theory of planned behavior is reviewed. Next the research model is presented, the methodology is described, and the results are analyzed. Finally, the findings are discussed, along with implications for economists and future avenues for research are presented.

LITERATURE REVIEW

Neoclassical economic theory assumes "bounded rationality," meaning that individuals almost always weigh their opportunity costs and choose an action that will increase their utility. Only occasionally will individuals make impulse decisions. Fishbein and Ajzen's (1975) theory of reasoned action predicts that subjective norms and attitudes are good predictors of intent, which in turn predicts behavior. Sheppard et al. (1988) analyzed 86 Theory of Reasoned Action studies, finding an average correlation of over 0.53 between intention and behavior. Relying on this work, the correlation between intent and action is acknowledged, but not tested, here. The theory of reasoned action evolved into the theory of planned behavior, which adds self-efficacy as a cause of intent (Ajzen, 1985 and Ajzen, 1991). This paper compares the relationships of one traditional dependent variable, intent to act, during a global financial crisis according to the theory of planned behavior, as adapted for the specifics of this financial crisis. Additionally, we control for standard demographic variables, which we expect to have no significant effect.

HYPOTHESES AND MODEL DESIGN

Intent to Change Jobs and Intent to Move Money

Intent is the extent to which a person is willing to exert an effort in order to perform a specified behavior (e.g. changing jobs). This paper measures intent to react to the national financial crisis by changing income streams (voluntary employment change) and investment allocation. Respondents were asked for example, on a 5-point scale how true (1= very untrue and 5=very true) was the following statement: "... I intend to move my financial assets from financial markets to cash or "... I intend to move my financial assets from financial markets into banks. The five point scale remains constant for all hypotheses.

Primary Dependent Variables

Perceived behavioral control is the amount of effect that people believe they have on their financial circumstances. A person may want to change jobs, but feel that there are no comparable jobs available. Stated as a hypothesis, perceived behavioral control is expected to have a significant, positive effect on both intents, or:

[H.sub.1]: Int Job = [B.sub.0] + [B.sub.1] * PBC

[H.sub.2]: Int Invest = [B.sub.0] + [B.sub.1] * PBC

Where Int Job is the intent to change jobs, Int Invest is the intent to change one's investment portfolio to a more conservative mix of savings and other insured investments, and PBC is perceived behavioral control over one's financial situation.

Ajzen (1991) found that awareness of other people's opinions produced changes in respondents' intents. Subjective norms are defined here as "the awareness of peers' changing asset allocations (jobs)." Applied to this study, the general construct of subjective norms will be tested to see if significant others' opinions and purported actions affect peoples' intent to change jobs or reallocate investments. Two measures are: "As a result of current changes in the economy my relatives are moving their financial assets from financial markets into banks" and "As a result of the current changes in the economy my relatives are moving their financial assets from financial markets into cash." Consistent with the theory of planned behavior, it is anticipated that the relationship between subjective norms and both intents is positive and significant:

[H.sub.3]: Int Job = [B.sub.0] + [B.sub.1] * NORM

[H.sub.4]: Int Invest = [B.sub.0] + [B.sub.1] * NORM

Where NORM measures subjective norms, which is how the respondents' friends and family are reacting to the crisis in terms of moving jobs and making their portfolio more conservative.

Ajzen (1991) tested the effect of self-efficacy, which is the amount of confidence one has in his/her own abilities. Consistent with the theory of planned behavior, it is anticipated that the relationship between self-efficacy and both intents is positive and significant:

[H.sub.5]: Int Job = [B.sub.0] + [B.sub.1] * SE

[H.sub.6] : Int Invest = [B.sub.0] + [B.sub.1] * SE

Where self-efficacy is the confidence one has in his/her own ability to change jobs or to make his/her portfolio mix to more conservative savings accounts.

Ajzen (1991) also tested the effect of affective attitude on intent, finding a significant positive relationship. Attitude can be generally defined as "how favorably or unfavorably the examined behavior is viewed." Attitude is operationalized as participants' responses to survey questions on how secure they felt about three aspects of their finances: savings accounts, investment funds (stocks and bonds) and incomes from their jobs. Respondents were asked to indicate, for example, how true the following statement was: "I feel that my savings in a bank is secure." It is anticipated that the relationship between attitude about the economy and both intents is negative and significant:

[H.sub.7]: Int Job = [B.sub.0] - [B.sub.1] * ATT

[H.sub.8]: Int Invest = [B.sub.0] - [B.sub.1] * ATT

[FIGURE 1 OMITTED]

Control variables (including age, gender, household income, racial identity, religiosity and experience) were also tested, with no significant results expected. The model can be expressed pictorially, as shown in Figure 1.

Sample and Data Collection

Approximately 458 members of a South Texas university's students, faculty members, and administrators/staff participated in this survey. Respondents from each of the categories were selected both purposively and on the basis of convenience. For example, those professors teaching classes of over 60 students were more likely to be solicited for permission to administer the questionnaires in their classes than those with smaller classes. Results for students did not vary significantly from the results of faculty and staff, indicating the fitness of students as subjects. Care was taken to ensure that participating students came from different class standings (freshmen, to graduate) and that faculty and staff from each college in the university was represented. Overall, the sample reflects the general demographic distribution of the university. Unlike students' questionnaires, however, faculty members, staff and administrators' questionnaires and Informed Consent Forms were mailed with separate return, self-addressed envelopes.

To explore any possible bias resulting from the use of students, bivariate correlations between demographic data and the independent variables (perceived behavioral control, norms, self-efficacy and attitude) and dependent variables (intent to change job, intent to reallocate investments) were calculated. There were no significant correlations, except as noted in the results section. Based on these results, it appears that demographic factors are generally not significant in explaining intent; therefore, the use of student subjects, whose demographic data may not be reflective of the general population, can provide useful information.

The survey instrument itself was extensive and collected information beyond that pertaining to the Theory of Planned behavior and control variables. Only information pertaining to those constructs was extracted and analyzed here. The survey is shown in Appendix A. Note that some questions are reverse-scaled to protect against positive response bias. Written instructions were included with the instrument to the participants, to assure the confidentiality of participants and stress the voluntary nature of participation.

The strength of the model and the scales used to measure their underlying latent constructs, shown in Figure 1, were assessed by applying partial least squares (PLS) analysis. PLS addresses both the effectiveness of the model and the reliability of the underlying measures simultaneously and has many additional advantages, such as relaxed error and distribution assumptions (Wold, 1982).

RESULTS

The age of the participants ranged from 16 to 71, with a median age of 23. Fifty-nine percent were female. Respondents included those with very little perceived experience to those with more extensive experience. The average participant rated herself as having experience of 3.0 on a 5-point scale. Approximately 71 percent of the respondents live in households with monthly income of at least $2,000, and the average monthly income was $4,818, similar to that of the national average.

In order to assess the construct validity of each measurement item, factor loadings are calculated. A factor loading of 0.70 or greater is considered to be a substantial correlation between the indicator and the latent variable (Chin, 1998). Barclay et al. (1995) recommend a loading of 0.707 or higher but he notes that it is not uncommon for items in newly developed scales to fail to meet the .707 level of reliability. Raubenheimer (2004) uses 0.40 for central factors and 0.25 for other factors. Because PLS minimizes the error variance for the whole model, newly developed scale items will generally be weighted less.

The self-efficacy factors did not hold together well, and those items that did not load well with others in the group were correspondingly weighted very low. Items for the other factors, with only five exceptions, have factor loadings of 0.70 or greater. All but one of these exceptions was greater than 0.60 (see Table 1).

The results of the confirmatory factor analysis suggest that the measurement items within each scale are highly correlated with the underlying latent variable. Additionally, 0.50 or more of the average variance for each factor is explained as required by Chin (1998) and Hock and Ringle (2006), with the exception of the self-efficacy construct. This indicates that the measurement items in these scales exhibit convergent validity, in that they are highly correlated to each other due to a single underlying construct. The average variance explained by the indicators is summarized in Table 2, with cells of 0.50 or more shaded.

To test the reliability of each of the scales, a composite reliability is also presented in Table 2. Except for the self-efficacy construct, each of the reliability statistics generally approaches or exceeds the 0.80 recommended by Nunnally and Bernstein (1994), and exceed 0.60 used by Chin (1998) and Hock and Ringle (2006). These cells are shown as shaded in the table.

The correlations among the latent variables are shown in Table 3, with the numbers presented in the diagonal depicting the square root of the average common variance extracted by the measurement items within the scale (the average inter-item correlation). The correlations among the latent variables are smaller than the square root of the common variance extracted within each scale, demonstrating divergent validity (items within a scale are more significantly related to one another than to items in other scales). Based on the preceding results, the measurements exhibit reasonable validity and reliability.

The path coefficients to the indicators from the latent variables (epistemic correlations) are presented in Figure 2. Three path coefficients are significant at [alpha] < 0.05 and of the correct sign, supporting hypotheses 3 (Norm to Intent to Move Jobs), 4 (Norm to Intent to Reallocate Assets) and 7 (Attitude to Intent to Move Jobs). All other paths (hypotheses 1, 2, 5, 6 and 8) were insignificant.

[FIGURE 2 OMITTED]

DISCUSSION

A path coefficient greater than 0.20 is meaningful per Chin (1998). Analyses of the results show that norms are a large, significant determinant of whether people intend to reallocate their assets.

Attitude is the significant determinant of whether people intend to move jobs. Norms influence whether one intends to change jobs, but not as much as attitude. The theory of planned behavior appears to be only moderately useful in predicting job turnover in times of financial crisis.

The amount of variance in the endogenous variables explained by the model is represented by the squared multiple correlations of 0.135 for intent to move jobs and 0.364 for intent to reallocate assets. Per Chin (1998) and Hock and Ringle (2006), an R-squared of 0.67 is considered substantial, 0.33 is moderate and 0.19 is weak explanatory power for dependent variables. The model appears to explain a substantial portion of the variance in intent to reallocate money. To determine the usefulness of the research model in Figure 2, the results of this model are compared to those from a simple model, in which norms are the only antecedents to intent to move money in to safer investment vehicles. In the simple model, the path from norms to intent to move money is significant ([alpha] [less than or equal to] 0.005), the explained variance in the attitude variable is 0.351 and the path size is 0.592. The addition of other independent variables do not add much explanatory power to the model, indicating that in predicting whether people will move out of the stock market and into conservative bank accounts, people are most heavily influenced by the behavior of their peers (norms). They make their investment decisions by following the crowd. Over one-third of people's investment decisions in a crisis come from referencing the behavior of family and friends, a result that lends credence to the powerful intrusion of social psychology on the otherwise rational man (homo economis).

Indeed, a blended, behavioral economics approach is gathering favor in policy-setting circles (Spiegel, 2009). To test whether the participants are influential on their family and friends rather than the other way around, the model was revised to show causality in the opposite direction and re-tested. The result was significantly worse. It appears that at least with respect to norms, people are following the crowd, not leading it, consistent with the theory of planned behavior, and encouraging a deeper look at collective economic behavior through a social psychology lens.

FURTHER RESEARCH

The effect of norms on individuals' decisions to move money dominates the findings in this paper. This information is useful and simultaneously consistent with behavioral economic theory and contrary to economic theory portraying each individual investor as a "rational man." Much of the recent behavioral economic theory centers on how individuals behave. From these findings, social psychology theory might deserve a second, harder look. Why do people follow the crowd? Economically, how do crowds behave?

Further, if people are following the crowd when making decisions, how should policy makers respond? Should popular opinion alone rule, and if so, should we (how can we) influence the popular opinion in times of economic crisis?

LIMITATIONS

The self-efficacy construct was measured essentially with a single item scale, in that the being a partner in a business modeled well with the theory of planned behavior, but the other measures of independence did not. It is preferable that measurement scales contain multiple, cohesive items. Future research with improved self-efficacy measures might lead to interesting and significant findings.

Finally, actual behavior was not included in the study. This is not a substantial problem because previous studies in the behavioral intentions research stream have supported a strong relationship between intention and actual behavior.

CONCLUSION

In predicting people's intent to change jobs, our model was weak, but with some significant findings: we find evidence that norms and attitude toward conservative financial investment strategies drive peoples' intent to change jobs. In predicting people's intent to move their money to conservative investments, like bank accounts, the model is much more robust, with over 36% of the intent explained by the model. Norms are significant and strongly positive. People intended to react to the global financial crisis the same way their peers did, indicating a strong social aspect to individuals' plans to handle their personal finances. This finding is important, adding to the growing literature that people are social, not strictly rational investors.

APPENDIX

Today's date

Read each item and, as honestly as you can, answer the question: "How
characteristic or true is this of me?" Circle the appropriate number,
using the following scale:

1 = very untrue   2 = untrue   3 = neutral   4 = true   5 = very true
DK = don't know

1.    I believe that getting together with one's         1 2 3 4 5
      friends to party is one of life's important
      pleasures.

2.    Familiar childhood sights, sounds, and smells      1 2 3 4 5
      often bring back a flood of wonderful memories.

3.    Fate determines much in my life.                   1 2 3 4 5

4.    I often think of what I should have done           1 2 3 4 5
      differently in my life.

5.    My decisions are mostly influenced by people and   1 2 3 4 5
      things around me.

6.    I believe that a person's day should be planned    1 2 3 4 5
      ahead each morning.

7.    It gives me pleasure to think about my past.       1 2 3 4 5

8.    I do things impulsively.                           1 2 3 4 5

9.    If things don't get done on time, I don't worry    1 2 3 4 5
      about it.

10.   When I want to achieve something, I set goals      1 2 3 4 5
      and consider specific means for reaching those
      goals.

11.   On balance, there is much more good to recall      1 2 3 4 5
      than bad in my past.

12.   When listening to my favorite music, I often       1 2 3 4 5
      lose all track of time.

13.   Meeting tomorrow's deadlines and doing other       1 2 3 4 5
      necessary work come before tonight's play.

14.   Since whatever will be will be, it doesn't         1 2 3 4 5
      really matter what I do.

15.   I enjoy stories about how things used to be in     1 2 3 4 5
      the "good old times."

16.   Painful past experiences keep being replayed in    1 2 3 4 5
      my mind.

17.   I try to live my life as fully as possible, one    1 2 3 4 5
      day at a time.

18.   It upsets me to be late for appointments.          1 2 3 4 5

19.   Ideally, I would live each day as if it were my    1 2 3 4 5
      last.

20.   Happy memories of good times spring readily to     1 2 3 4 5
      mind.

21.   I meet my obligations to friends and authorities   1 2 3 4 5
      on time.

22.   I've taken my share of abuse and rejection in      1 2 3 4 5
      the past.

23.   I make decisions on the spur of the moment.        1 2 3 4 5

24.   I take each day as it is rather than try to plan   1 2 3 4 5
      it out.

25.   The past has too many unpleasant memories that I   1 2 3 4 5
      prefer not to think about.

26.   It is important to put excitement in my life.      1 2 3 4 5

27.   I've made mistakes in the past that I wish I       1 2 3 4 5
      could undo.

28.   I feel it's more important to enjoy what you're    1 2 3 4 5
      doing than to get work done on time.

29.   I get nostalgic about my childhood.                1 2 3 4 5

30.   Before making a decision, I weigh the costs        1 2 3 4 5
      against the benefits.

31.   Taking risks keeps my life from becoming boring.   1 2 3 4 5

32.   It's more important for me to enjoy life's         1 2 3 4 5
      journey than to focus only on the destination.

33.   Things rarely work out as I expected.              1 2 3 4 5

34.   It's hard for me to forget unpleasant images of    1 2 3 4 5
      my youth.

35.   It takes joy out of the process and flow of my     1 2 3 4 5
      activities if I have to think about goals,
      outcomes, and products.

36.   Even when I am enjoying the present, I am drawn    1 2 3 4 5
      back to comparisons with similar past
      experiences.

37.   You can't really plan for the future because       1 2 3 4 5
      things change so much.

38.   My life path is controlled by forces I cannot      1 2 3 4 5
      influence.

39.   It doesn't make sense to worry about the future,   1 2 3 4 5
      since there is nothing that I can do about it
      anyway.

40.   I complete projects on time by making steady       1 2 3 4 5
      progress.

41.   I find myself tuning out when family members       1 2 3 4 5
      talk about the way things used to be.

42.   I take risks to put excitement in my life.         1 2 3 4 5

43.   I make lists of things to do.                      1 2 3 4 5

44.   I often follow my heart more than my head.         1 2 3 4 5

45.   I am able to resist temptations when I know that   1 2 3 4 5
      there is work to be done.

46.   I find myself getting swept up in the excitement   1 2 3 4 5
      of the moment.

47.   Life today is too complicated; I would prefer      1 2 3 4 5
      the simpler life of the past.

48.   I prefer friends who are spontaneous rather than   1 2 3 4 5
      predictable.

49.   I like family rituals and traditions that are      1 2 3 4 5
      regularly repeated.

50.   I think about the bad things that have happened    1 2 3 4 5
      to me in the past.

51.   I keep working at difficult, uninteresting tasks   1 2 3 4 5
      if they will help me get ahead.

52.   Spending what I earn on pleasures today is         1 2 3 4 5
      better than saving for tomorrow's security.

53.   Often luck pays off better than hard work.         1 2 3 4 5

54.   I think about the good things that I have missed   1 2 3 4 5
      out on in my life.

55.   I like my close relationships to be passionate.    1 2 3 4 5

56.   There will always be time to catch up on my        1 2 3 4 5
      work.

57.   In Fall 2007 I felt that my savings in a bank      1 2 3 4 5 DK
      were secure.

58.   In Fall 2007 I felt that my investment funds       1 2 3 4 5 DK
      (stocks & bonds) were secure.

59.   In Fall 2007 I felt that my job (source of         1 2 3 4 5 DK
      income) was secure.

60.   In August 2008 I felt that my savings in a bank    1 2 3 4 5 DK
      were secure

61.   In August 2008 I felt that my investment funds     1 2 3 4 5 DK
      (stocks & bonds) were secure.

62.   In August 2008 I felt that my job (source of       1 2 3 4 5 DK
      income) was secure.

63.   Today I feel that my savings in a bank is          1 2 3 4 5 DK
      secure.

64.   Today I feel that my investment funds (stocks &    1 2 3 4 5 DK
      bonds) are secure.

65.   Today I feel that my job (source of income) was    1 2 3 4 5 DK
      secure.

66.   As a result of changes in the economy many of my   1 2 3 4 5 DK
      relatives are moving their financial assets from
      financial markets into banks.

67.   As a result of changes in the economy many of my   1 2 3 4 5 DK
      relatives are moving their financial assets from
      financial assets into cash.

68.   As a result of changes in the economy many of my   1 2 3 4 5 DK
      relatives are looking for a new job

69.   As a result of changes in the economy many of my   1 2 3 4 5 DK
      relatives are retiring.

70.   As a result of changes in the economy many of my   1 2 3 4 5 DK
      relatives are training for a new job.

71.   As a result of changes in the economy many of my   1 2 3 4 5 DK
      relatives are (please specify and state extent
      to which it true.

72.   As a result of how I feel now, I intend to move    1 2 3 4 5 DK
      my financial assets from financial markets into
      banks.

73.   As a result of how I feel now, I intend to move    1 2 3 4 5 DK
      my financial assets from financial assets into
      cash.

74.   As a result of how I feel now, I intend to look    1 2 3 4 5 DK
      for a new job.

75.   As a result of how I feel now, I intend to         1 2 3 4 5 DK
      retire.

76.   As a result of how I feel now, I intend to train   1 2 3 4 5 DK
      for a new job.

77.   As a result of how I feel now, I intend to         1 2 3 4 5 DK
      (please specify and state extent to which it. is
      true.)

78.   I have the power to improve my current financial   1 2 3 4 5 DK
      situation.

79.   I understand what is going on in the economy       1 2 3 4 5 DK

80.   I understand what is going on in the financial     1 2 3 4 5 DK
      markets

DEMOGRAPHICS: Please circle the number that corresponds to the category that best describes you:

Sex:

1. Male

2. Female

Age at last birthday --

Zip code --

I own my own business.

1. Yes

2. No

I'm a partner in a business.

1. Yes

2. No

I do independent consulting work.

1. Yes

2. No

I work in -- industry

Currently taking college classes?

1. Yes

2. No

Your major (college students only) --

Current household monthly income (approximately) --

Approximate dollar value of your financial assets (savings, investments etc.)?

1. less than 25,000

2. 25,000-49,999

3. 50,000-74,999

4. 75,000-99,999

5. 100,000-124,999

6. 125,000-149,999

7. 150,000-174,999

8. 175,000-199,999

9. 200,000-224,999

10. 225,000-249,999

11. 250,000-274,999

12. 275,000-299,999

13. 300,000-324,999

14. 325,000-349,999

15. 350,000-374,999

16. 375,000-399,999

17. 400,000-424,999

18. 425,000-449,999

19. 450,000-474,999

20. 475,000-499,999

21. 500,000 +

IN PERCENTAGES, how your financial assets are distributed among the following (must add up to 100%).

1. Checking accounts --

2. Savings accounts --

3. Stocks/bonds/mutual funds --

4. Retirement/pension funds --

5. Other -- (please specify --)

IN PERCENTAGES, how your real assets are distributed among the following (must add up to 100%).

1. Home --

2. Vehicles --

3. Other real estate --

4. Personal property (furniture, tools electronics, jewelry, etc.) --

5. Other -- (please specify: --)

Highest level of educational attainment:

1. Less than high school

2. High school/GED

3. Some college

4. Bachelors degree

5. Masters degree

6. Above Masters degree

What is the subject area is your highest degree (college graduates only)? --

I would classify my business experience level as:

1. Very Low

2. Low

3. Average

4. High

5. Very High

What is your Racial/ethnic identity?

1. African American

2. Asian American

3. Hispanic American

4. Native American

5. White American

What is your religious affiliation?

1. Catholic

2. Protestants (all Christian denominations that are not Catholic)

3. Jewish

4. Moslem

5. Atheist

6. Other (please specify) --

How many times do you pray (on your own) weekly ? --

How many times do you attend a religious activity (church etc)? --

How important is religion in your personal decisions?

1. Very unimportant

2. Unimportant

3. Important

4. Very Important

How would you describe yourself politically?

    Very Liberal   Moderate   Very Conservative
1           2    3    4     5 6

REFERENCES

Ajzen, I. (1985). "From Intentions to Actions: A Theory of Planned Behavior," Action Control--From Cognition to Behavior, J. Kuhl and J. Beckmann (eds.), (Berlin: Springer-Verlag), 11-39.

Ajzen, I. (1991). "The Theory of Planned Behavior," Organizational Behavior and Human Decision Processes, Vol. 50, 179-211.

Barclay, D., R. Thompson, C. Higgins (1995). The partial least squares (PLS) approach to causal modeling: Personal computer adoption and use as an illustration. Technology Studies, 285-323.

Chin, W.W. (1998). "The Partial Least Squares Approach for Structural Equation Modeling," In Modern Methods for Business Research, G.A. Marcoulides, Ed. (Lawrence Erlbaum Associates, Mahwah, NJ).

Fishbein, M., and Ajzen, I. (1975). Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research, (Addison-Wesley, Reading, MA).

Nunally, J.C., and Bernstein, I.H. (1994). Psychometric Theory, (McGraw-Hill, New York), 3rd Edition.

Sheppard, B.H., Hartwick, J., and Warshaw, P.R. (1988). "The Theory of Reasoned Action: A Meta-Analysis of Past Research with Recommendations for Modifications and Future Research," Journal of Consumer Research, Vol. 15 (December), 325-343.

Spiegel, A. (2009). "Using psychology to save you from yourself." Nation, June 10, 2009. www.npr.org/templates/story/story.php?storyId=104803094, retrieved 6/10/2009.

Taylor, S., and Todd, P. (1995). "Understanding Information Technology Usage: A Test of Competing Models," Information Systems Research, (June), 144-176.

Wold, H. (1982). Soft modeling, the basic design and some extensions.

Valrie Chambers, Texas A&M University-Corpus Christi

Bilaye R. Benibo, Texas A&M University-Corpus Christi

Marilyn Spencer, Texas A&M University-Corpus Christi

Table 1 Primary Measurement Model Variables Using Primary Least
Squares

Factor                          Indicator/    Factor    Weight
                                Question #    Loading

Intent--Reallocate Assets           71        0.8223    0.4987
Intent--reallocate Assets           72        0.9114    0.6883
Intent--Move Job                    73        0.8667    0.6014
Intent--Move Job                    74        0.1645   -0.0237
Intent--Move Job                    75        0.8586    0.5809
Perceived Behavioral Control        76        0.7382    0.5844
Perceived Behavioral Control        77        0.8315    0.3199
Perceived Behavioral Control        78        0.7732    0.4238
Self-efficacy                  Own Business   0.3674    0.0946
Self-efficacy                    Partner      0.9846    0.9898
Self-efficacy                  Independent    0.0561   -0.1663
Norms                               66        0.6042    0.6220
Norms                               67        0.6315    0.3581
Norms                               68        0.7574    0.3107
Norms                               69        0.6986    0.2594
Norms                               70        0.8183    0.3813
Attitude                            63        0.7671    0.3849
Attitude                            64        0.6312    0.2104
Attitude                            65        0.8587    0.7355

* shaded cells are high enough (0.40 or 0.25) for exploratory research
per Raubenheimer (2004).

Table 2 Common Variance Explained and Composite Reliability Measures

Construct                     Average Variance    Composite
                                 Explained       Reliability

Intent--Reallocate Assets          0.753            0.859
Intent--Move Job                   0.505            0.706
Perceived Behavioral Control       0.611            0.825
Self-efficacy                      0.369            0.512
Norms                              0.499            0.831

Table 3--Correlations among Latent Variables

Construct                      Intent to   Intent to   Perceived
                               Relocate    Move Job    Behavioral
                                Assests                 Control

Intent--Reallocate Assets        0.868
Intent--Move Job                 0.19        0.711
Perceived Behavioral Control    -0.08        0.071        0.782
Self-efficacy                    0.124       0.023        0.02
Norms                            0.592       0.203       -0.081
Attitude                        -0.231      -0.315        0.14

Construct                       Self-     Norms   Attitudes
                               efficacy

Intent--Reallocate Assets
Intent--Move Job
Perceived Behavioral Control
Self-efficacy                   0.607
Norms                           0.162      0.706
Attitude                        0.049     -0.21     0.758

* The numbers presented in the diagonal depicting the square root of
the average common variance extracted by the measurement items within
the scale.