Innate: Development of a New Intrinsic Motivation Measure Using Confirmatory Factor Analytic Methods

ABSTRACT - That component of motivation which is self-generated has been of interest to marketing researchers for a long time, and has been measured in different ways. This paper traces the construct of intrinsic motivation back to its origins in the socio-psychological literature, and develops a new instrument to measure it. The scale development process employed includes the use of confirmatory factor analysis to demonstrate unidimensionality and discriminant validity. It is believed that this measure could be adapted to explain variance in the behavior of telemarketing, customer service or other service employees in addition to that of salespeople.

Citation:

M. V. Thakor (1994) ,"Innate: Development of a New Intrinsic Motivation Measure Using Confirmatory Factor Analytic Methods", in AP - Asia Pacific Advances in Consumer Research Volume 1, eds. Joseph A. Cote and Siew Meng Leong, Provo, UT : Association for Consumer Research, Pages: 116-121.

Asia Pacific Advances in Consumer Research Volume 1, 1994      Pages 116-121

INNATE: DEVELOPMENT OF A NEW INTRINSIC MOTIVATION MEASURE USING CONFIRMATORY FACTOR ANALYTIC METHODS

M. V. Thakor, Concordia University

ABSTRACT -

That component of motivation which is self-generated has been of interest to marketing researchers for a long time, and has been measured in different ways. This paper traces the construct of intrinsic motivation back to its origins in the socio-psychological literature, and develops a new instrument to measure it. The scale development process employed includes the use of confirmatory factor analysis to demonstrate unidimensionality and discriminant validity. It is believed that this measure could be adapted to explain variance in the behavior of telemarketing, customer service or other service employees in addition to that of salespeople.

INTRODUCTION

Motivation has long been studied in marketing, particularly in the salesperson literature (Churchill, Ford and Walker 1979) which has been concerned with how it is affected by the work environment and how if affects, in turn, job outcomes such as performance (Walker, Churchill and Ford 1977). Motivation-based research in marketing has however operated from diverse theoretical contexts. Researchers working within the expectancy theory framework have categorized some rewards as internally mediated in that they pertain to higher order needs and are attained by the employee herself, in contrast to externally mediated rewards such as pay and promotion which can be provided by the firm (Cron, Dubinsky and Michaels 1988). Similar to the conceptualization of this aspect of motivation as the valence attached to certain rewards, some researchers working from an attributional perspective have called this an "intrinsic reward orientation" (Sujan 1986, p.42). Overall, however, considerable ambiguity still surrounds both the conceptualization of intrinsic motivation and its measurement, with no single approach having emerged as a standard. The extant measures also do not seem well grounded theoretically, or to have considered the socio-psychological literature where this construct has a long provenance.

Theorists in areas such as service marketing, as well as sales management issues, have suggested that employees must not just meet but exceed customer expectations, or "delight" customers (Berry and Parasuraman 1991). This underscores the necessity for employees to take pride in their jobs and see them as much more than simply a meal ticket. The recent emphasis on salespeople's customer orientation (Saxe and Weitz 1982; Michaels and Day 1985) also draws attention to an area where measurement of intrinsic motivation would presumably be of some importance, because of the limits to reward systems in mandating a long-term view or excellence in customer service. If intrinsically motivated personnel are indeed crucial to the success of many firms in highly competitive industries, there appears to be a clear need for a theory-based measure of intrinsic motivation with sound psychometric properties. The present paper works within the scale development paradigm recommended by Churchill (1979) and updated by Gerbing and Anderson (1982) to develop such a measure. The measure is tested on a sample of 281 respondents using a variety of techniques including confirmatory factor analysis.

DEFINITIONS OF INTRINSIC MOTIVATION

Deci (1975, 1980) defined intrinsic motivation as caused by the underlying need for a sense of competence and self-determination, with self-determination being the more fundamental of the two. Although these two often covary in real-life situations, self-determination is conceptualized to be more fundamental because the acquisition of competence has been found not to be motivating unless it occurs within the context of self-determination. Deci and Ryan (1980) provide an operational definition of intrinsically motivated behaviors as "those that are performed in the absence of any apparent external contingency". In line with this operationalization, intrinsic motivation has typically been measured in the social psychology literature by observing behavioral persistence in a free-choice period following the removal of extrinsic rewards.

In the marketing literature, Tyagi (1985) took an expectancy value approach to the measurement of intrinsic motivation, and assessed the valence of six "intrinsic outcomes" for salespeople using a thermometer scale. The outcomes were decided upon, and items relating to each of them created, after interviewing sales managers and salespeople. Although intrinsic motivation is not clearly defined, Tyagi refers to job characteristics such as challenge and variety resulting in the job having "high personal meaning" for the individual, and characteristics such as task identity as producing "feelings of doing a worthwhile job". He goes on to say: "Such feelings or beliefs are what result in intrinsic motivation. A job offering considerable autonomy is also likely to induce intrinsic motivation". In more recent studies, Sujan (1986) and Spiro and Weitz (1990) follow Dyer and Parker (1975) and define intrinsic motivation as the motivation to seek "rewards derived directly from or inherent in the task or job itself-associated with the content of the task or job".

There appear to be substantial differences between the marketing and the socio-psychological literatures on the issue of how intrinsic motivation should be conceptualized. In particular, the definitions used in the marketing literature do not appear to be grounded in fundamental processes underlying intrinsic motivation such as the need for competence and self-determination. The Dyer and Parker definition of intrinsic motivation antedates most of the work on intrinsic motivation by Deci and others, and fails to specify precisely which rewards are being sought from the task by the person who is intrinsically motivated, whereas it is clear from the writings of Deci, among others, that what is being sought is a sense of competence. Similarly, the Tyagi definition seems deficient in that the effects of such job characteristics as challenge and variety on the individual's intrinsic motivation are probably mediated by the increased feelings of competence experienced by the individual in coping with the job's variety and challenge.

The conceptual definition of intrinsic motivation found in the socio-psychological literature (Deci 1975; Deci and Ryan 1980; Deci and Porac 1978; Ryan Mims and Koestner 1983) not only appears sound, but also enjoys considerable empirical support, and hence will be the conceptualization followed in this study. We define intrinsic motivation as that drive born out of a need for competence and self-determination which leads individuals to behave in ways other than those strictly compelled by external contingencies.

PREVIOUS MEASURES OF INTRINSIC MOTIVATION

As noted earlier, researchers in the socio-psychological literature, where most of the studies of intrinsic motivation have been conducted, have relied largely on an experimental paradigm, and have measured the subject's intrinsic motivation as the amount of time the subject spent working on the experimental task after the removal of external contingencies. This is usually referred to as a "free choice" measure of intrinsic motivation. A few socio-psychological researchers, have, however, tried to use self-report measures in addition to free choice measures. Ryan Mims and Koestner (1983) report using an 11-item self-report measure of intrinsic motivation, where the items all related to interest in, enjoyment of, and attention paid to the task. No psychometric properties are reported, however, nor are the items themselves. They did report that when scores on all items were averaged and correlated with the free choice measure also used in their study, a correlation of 0.42 (p < .001) was obtained.

Other research in the psychological literature employing both self-report and free-choice measures has produced mixed results. The correlation found by Ryan et al. is very similar in magnitude to that obtained by Harackiewicz (1979), who also used both free choice and self report measures (the items in the self-report measure were not listed). Luyten and Lens (1981) in their study found no significant correlation between free choice and self report measures, but this may be due to measurement error, since they used a single-item self report measure. It seems that self-report measures of intrinsic motivation as used in the psychology literature have had some success in replicating intrinsic motivation scores obtained through free choice measures. It is notable however that free-choice measures of intrinsic motivation as used in socio-psychological experiments do not permit determination of the reasons why subjects continue to work on a task after removal of external contingencies, and consequently do not allow a test of the conceptualization of intrinsic motivation in terms of the individual's need for competence and self-determination. It may be that subjects continue with a task after removal of external contingencies because of simpler motivations such as an interest in the task, rather than the fulfilment of any other needs, and in fact some researchers, such as Kruglanski (1978), speak of intrinsic motivation to perform an activity being manifested in "interest in, enjoyment of, and freedom at (sic) the activity".

In the marketing literature, Spiro and Weitz (1990) measured intrinsic motivation using a 6-item scale with items such as "selling is not fun" (reversed), "selling a customer is like playing a game", and "interacting with customers is exciting and challenging". This scale has a Cronbach's alpha of 0.79. Tyagi (1985) takes a different approach and develops items relating to "intrinsic outcomes", but there appears to have been no a priori basis for selection of these outcomes. It appeared that the items used by these researchers did not necessarily connect with the underpinnings of intrinsic motivation as conceptualized in the socio-psychological literature. It was felt to be best to approach the measurement of intrinsic motivation with a combination of items tapping the respondent's interest in and enjoyment of the activity, as well as items addressing the issue of whether the respondent experiences, and feels rewarded by, enhanced feelings of competence while performing the activity.

SCALE DEVELOPMENT PROCEDURE

A promising starting point for the measurement of the intrinsic motivation construct was thought to be with the job cognitions scale developed by Brief and Roberson (1987) based on the job outcomes identified by Dawis and Lofquist (1984). The items in this scale have been shown to have adequate internal consistency reliability, with a coefficient alpha of 0.73 being reported by Brief and Roberson (1987) and one of 0.83 being reported by Atieh (1987). Williams (1988) showed using exploratory factor analysis that a two factor model proposing intrinsic and extrinsic job outcomes was supported as postulated a priori based on Dawis and Lofquist (1984). The eigenvalues for the first two factors were 5.1 and 2.1 respectively, and 35% of the variance in the data was explained by these two factors, with eigenvalues for other factors falling off sharply. Only four of twenty items relating to job outcomes did not load on these two factors, and the remaining sixteen items loaded highest on the appropriate factor. Based on these results, Williams combined 8 items each into a Job Cognitions-Extrinsic and a Job Cognitions-Intrinsic scale. The scales had a 0.42 correlation, and coefficient alphas of 0.83 and 0.72.

Some items from Williams' Job Cognitions-Intrinsic scale were used to measure salespeople's intrinsic motivation, and other items were added to complete the scale. An effort was made to customize the scale to the sales context, to ensure that the dimensions of "competence" and "self-determination" were tapped into, and to add a negatively worded item. Three of Williams' items were used-items no. 1, 2 and 3 of the scale used in the questionnaire and reported in Table 1 (all items required responses on a 7-point Likert scale anchored by "Strongly Disagree" and "Strongly Agree"). The fourth item, a negatively worded one, was added to counteract any acquiescence bias. The fifth item, "my job allows me to have control over my life", was intended to tap into the need for self-determination. The sixth item, "my job is exciting and challenging", is similar to one used by Spiro and Weitz. It was intended to tap into the competence dimension, as well as to assess if the job was interesting and enjoyable, per Kruglanski (1978). Finally, the seventh item, "my job allows me to grow and develop as a person", also tried to assess if the job was seen as promoting the competence level of the salesperson.

Several other items were included in the original list of items which, along with the construct definition, was discussed with several faculty members and sales professionals. Based on the feedback obtained, changes were made to eliminate ambiguous items as well as those that seemed to fall outside the defined domain of the construct. Finally, a pretest was conducted using about 38 sales professionals, and all the 7 items reported already. Quantitative analysis began at this point, with the calculation of means, standard deviations, a correlation matrix, coefficient alpha, item-to-total correlations and other descriptive statistics. The results seemed promising, especially in that correlations and item-to-total correlations were almost uniformly high. All 7 items were retained for use with the larger sample of industrial salespeople discussed below.

DATA

The scale was administered to alumni of a large Midwestern university who were employed as sales professionals. The overall response rate to the survey was 40.24%, which is considered excellent for mail surveys. Two hundred and eighty-one questionnaires from salespeople of industrial products, the target population for this study, were input to the scale analyses after some questionnaires were excluded because of incomplete responses or other reasons.

RESULTS

The items developed appeared to perform well in terms of their psychometric properties. The coefficient alpha for the full seven-item scale was 0.89, and exploratory factor analysis revealed a one-factor solution, with the factor accounting for 62.3% of the variance in the construct. Relevant statistics including means and standard deviations, and reliability analysis and exploratory factor analysis results can be found in Table 2. As will be seen, the items yielded a scale alpha of 0.89.

TABLE 1

INTRINSIC MOTIVATION ITEMS

TABLE 2

PRELIMINARY STATISTICS

TABLE 3

EXPLORATORY FACTORS ANALYSIS RESULTS

When subjected to exploratory factor analysis, one factor emerged with the eigenvalue corresponding to this factor being 4.35 and 62.3% of the variance being explained by it (see Table 3). Moreover, the next highest eigenvalue was only 0.79, so that either using the criterion of retaining only those factors with eigenvalues greater than 1 or by the criterion of a scree test, the results can be considered supportive of a one-factor solution.

In the next stage, all 7 items were subjected to a confirmatory factor analysis using LISREL VII in order to test unidimensionality. This produced a model that fit acceptably, with an adjusted goodness of fit index of 0.927, although the chi-square statistic was significant at 35.34 with 14 degrees of freedom due to its well-known sensitivity to sample size (Bentler 1990). The results are in Table 4. Despite the favorable indices of fit, the residuals were inspected for indications of potential problems (the large residuals are also in Table 4). The residuals between items 2, 6 and 7 may be due to the connections between the words growth, challenge and accomplishment (used in the three items) in our culture.

Overall, the scale seemed to be adequate for input to further analyses. The only question was about the suitability of item 5, because of the low squared multiple correlation coefficient associated with this item in the reliability analysis stage. On reflection, it was decided to retain item 5 in the scale so as to preserve the content validity of the scale, since this item is the only one which speaks of the empowering qualities of the job, an important aspect of our definition of intrinsic motivation. Item 7 was also retained despite the high residual associated with it because the ideas of growth and development contained in this item seemed important to the conceptualization of intrinsic motivation underlying this entire measurement effort.

TABLE 4

CONFIRMATORY FACTOR ANALYSIS

A further effort was undertaken to assess whether the scale discriminated sufficiently between three distinct constructs, which are however all related to the job context: organizational commitment, pay satisfaction and intrinsic motivation. The items commonly used to measure these constructs require respondents to provide affective reactions to some aspect of their work environment: their firm, their pay, or their work itself. Organizational commitment, was measured in this study using the full 15 item version of the Organizational Commitment Questionnaire developed by Porter et al. (1974), which has been used in over 90 studies to date (Mathieu and Zajac 1990). Pay satisfaction was measured using a 6 item scale of which 4 items were drawn from the INDSALES scale constructed by Churchill et al. (1974). These were slightly modified, and two other items added, in light of the considerable literature on the subject that has appeared since in the organizational behavior literature (Scarpello et al. 1988; Miceli and Lane 1990). The coefficient alpha for the OCQ in this study was 0.93, and that for the pay satisfaction scale was 0.89. While the foci of these three constructs is different, it might have been the case that an overall affective reaction to the work environment was responsible for the observed pattern of variances. It was consequently decided to run three different sets of confirmatory factor analyses, with two constructs being included in each set, per Anderson and Gerbing (1988). In the first instance, organizational commitment and intrinsic motivation were considered, and two confirmatory analyses were run, with the first one allowing the constructs to covary, and the second one constraining them to correlate perfectly. The logic of this is that "..a significantly lower ^c2 value for the model in which the trait correlations are not constrained to unity would indicate that the traits are not perfectly correlated and that discriminant validity is achieved" (Bagozzi and Phillips 1982). The results of this analysis, are reported along with those for the other pairs tested in Table 5.

As can be seen, the results provide support for the contention that an adequate level of discriminant validity between organizational commitment and intrinsic motivation has been achieved, since the chi-square statistic decreases significantly when they are allowed to covary, as compared to the model in which their intercorrelation is constrained to be one. Similarly, there is evidence for discriminant validity between organizational commitment and pay satisfaction, and between intrinsic motivation and pay satisfaction. It seems reasonable to conclude that the precautions taken during the item development and scale refinement stages to ensure discriminant validity were somewhat successful. Admittedly, the test employed here provides only weak evidence of discriminant validity, and it is recognised that a multi-trait multi-method approach would have to be employed in order to make claims of discriminant validity more confidently.

CONCLUSION AND SUGGESTIONS FOR FUTURE RESEARCH

It appears that the measure developed in this study constitutes a worthwhile tool for the study of several marketing issues. While the most obvious area of application is in sales force management, the scale should be usable in other job contexts, such as studies of the performance of telemarketing and customer service personnel, or that of service providers, or other employees in a firm. It will be noted that the items in the scale are not worded so restrictively as to be applicable only to salespeople.

While scale development was carried out here seeking to conform as closely as possible to the prescriptions found in the marketing and psychometric literatures, it is recognized that the establishment of construct validity (particularly nomological validity) is usually a slow process which depends on the accretion of evidence from many studies. Hopefully, this description of the measure's origins will encourage its use in studies spanning several content areas, contributing not only to its validation but also to the substantive body of marketing knowledge.

TABLE 5

DISCRIMINANT VALIDITY ASSESSMENT USING CFA

BIBLIOGRAPHY

Anderson James C. and David W. Gerbing (1988), "Structural Equation Modeling in Practice: A Review and Recommended Two-Step Approach", Psychological Bulletin, v. 103, 3, 411-423.

Atieh, J. The affective and cognitive components of job satisfaction. Unpublished dissertation.

Bagozzi Richard P. and Lynn W. Phillips (1982), "Representing and testing organizational theories: A holistic construal", Administrative Science Quarterly, 27, 459-489.

Bentler, P.M. (1990), "Comparative Fit Indexes in Structural Models", Psychological Bulletin, 107, 2, 238-246.

Berry, Leonard L. and A. Parasuraman (1991), Marketing Services, New York N.Y.: The Free Press.

Brief A. P. and Roberson L. (1987), "Job Attitude Organization: An Exploratory Study", paper presented at the National Meetings of the Academy of Management, New Orleans.

Dawis R. B. and Lofquist L. H. (1984), A Psychological Theory of Work Adjustment, Minneapolis: University of Minneapolis Press.

Churchill, Gilbert A. Jr., Neil M. Ford, and Orville C. Walker Jr. (1979), "Personal Characteristics of Salespeople And The Attractiveness of Alternative Rewards", Journal of Business Research, 7, 25-50.

Churchill, Gilbert A., Jr. (1979), "A Paradigm for Developing Better Measures of Marketing Constructs", Journal of Marketing Research, 16 (February), 64-73.

Cron, William L., Alan J. Dubinsky and Ronald E. Michaels (1988), "The Influence of Career Stages on Components of Salesperson Motivation", Journal of Marketing, 52 (January) 78-92.

Deci, Edward L. (1975), Intrinsic Motivation. New York: Plenum.

Deci, Edward L. (1980), The Psychology of Self-Determination. Lexington, Mass: D.C. Heath, Lexington Books.

Deci, Edward L. and Richard M. Ryan (1980), "The Empirical Exploration of Intrinsic Motivational Processes", in Advances in Experimental Social Psychology, Vol. 13, ed. L. Berkowitz, New York: Academic Press.

Deci, Edward L. and J. F. Porac (1978), "Cognitive Evaluation Theory and the Study of Human Motivation", in The Hidden Costs of Reward, eds. M. R. Lepper and D. Greene, Hillsdale NJ: Erlbaum.

Dyer Lee and Donald F. Parker (1975), "Classifying Work Outcomes in Motivational Research: An Examination of the Intrinsic-Extrinsic Dichotomy", Journal of Applied Psychology, 60 (August) 455-458.

Gerbing David W. and James C. Anderson (1982), "An Updated Paradigm for Scale Development Incorporating Unidimensionality and Its Assessment", Journal of Marketing Research, 25 (May), 186-192.

Harackiewicz, J. (1979), "The effects of reward contingency and performance feedback on intrinsic motivation", Journal of Personality and Social Psychology, 37, 1352-1363.

Kruglanski, Arie (1978), in The Hidden Costs of Reward, eds. M. R. Lepper and D. Greene, Hillsdale NJ: Erlbaum.

Luyten, H. and W. Lens (1981), "The effect of earlier experience and reward contingencies on intrinsic motivation", Motivation and Emotion, 5, 25-36.

Miceli, Marcia P. and Mathew C. Lane (1990), "Antecedents of Pay Satisfaction: A Review and Extension", in K. Rowland and G. Ferris (Eds.), Research in Personnel and Human Resources Management, vol. 9, Greenwich CT: JAI Press.

Michaels Ronald E. and Ralph L. Day (1985), "Measuring Customer Orientation of Salespeople: A Replication With Industrial Buyers", Journal of Marketing Research, 22 (November) 443-446.

Porter, L. W., R. M. Steers, R. T. Mowday, and P. V. Boulian (1974), "Organizational commitment, job satisfaction, and turnover among psychiatric technicians", Journal of Applied Psychology, 59, 603-609.

Ryan Richard M., Valerie Mims, and Richard Koestner (1983), "Relation of a Reward Contingency and Interpersonal Context to Intrinsic Motivation: A Review and Test Using Cognitive Evaluation Theory", Journal of Personality and Social Psychology, 45 (October), 736-50.

Saxe, Robert and Barton A. Weitz (1982), "The SOCO Scale: A Measure of the Customer Orientation of Salespeople", Journal of Marketing Research, 19 (August) 343-351.

Scarpello, Vida, Vandra Huber and Robert J. Vandenberg (1988), "Compensation Satisfaction: Its Measurement and Dimensionality", Journal of Applied Psychology, 73, No. 2, 163-171.

Spiro Rosann L. and Barton A. Weitz (1990), "Adaptive Selling: Conceptualization, Measurement, and Nomological Validity", Journal of Marketing Research, 27 (February) 61-69.

Sujan, Harish (1986), "Smarter Versus Harder: An Exploratory Attributional Analysis of Salespeople's Motivation", Journal of Marketing Research, 23 (February) 41-9.

Tyagi, Pradeep K. (1985), "Relative Importance of Key Job Dimensions and Leadership Behaviors in Motivating Salesperson Work Performance", Journal of Marketing, 49 (Summer) 76-86.

Walker, Orville C. Jr., Gilbert A. Churchill Jr., and Neil M. Ford (1977), "Motivation and Performance in Industrial Selling: Present Knowledge and Needed Research", 14 (May), 156-168.

Williams, Larry J. (1988), "Affective and Non-affective Components of Job Satisfaction and Organizational Commitment As Determinants of Organizational Citizenship and In-Role Behaviors", unpublished dissertation, Department of Management, School of Business, Indiana University, Bloomington 47405.

----------------------------------------