1 Introduction

A growing literature has demonstrated that spending money on life experiences contributes to greater well-being than spending money on material items (Caprariello and Reis 2010; Howell and Hill 2009; Howell et al. 2012; Nicolao et al. 2009; Van Boven and Gilovich 2003). However, recent studies suggest that individual differences moderate the relationship between experiential buying and well-being (Millar and Thomas 2009; Nicolao et al. 2009). Also, while experiential purchasers are often judged to be intrinsically motivated (Van Boven et al. 2010), it is possible that people spend money on life experiences for extrinsic motives because behavioral motives often operate independently from behavioral content (Sheldon et al. 2004). Thus, the impact of experiential purchases on well-being should be moderated by a person’s consumption motives. In order to test this proposition, our primary goal is to use self-determination theory (SDT; Deci and Ryan 1985) as an overarching framework to test for the relations between intrinsically and extrinsically motivated experiential buying and well-being. To accomplish this, the immediate research goal is to develop and validate a measure of motives for experiential consumption—the Motivation for Experiential Buying Scale (MEBS).

1.1 Experiential Consumption and Well-Being

Both philosophical adages and contemporary scholars have advocated for the consumption of experiences over material goods (Dunn et al. 2011; Fromm 1976; Kasser et al. 2004). Since Van Boven and Gilovich’s (2003) seminal article, the positive impact that experiential consumption (i.e., events or series of events that a person lives through)Footnote 1 has on hedonic well-being has been replicated across numerous samples and designs (Caprariello and Reis 2010; Howell and Hill 2009; Millar and Thomas 2009; Nicolao et al. 2009). Specifically, compared to material purchases, people are happier when thinking about their experiential purchases (Nicolao et al. 2009; Van Boven and Gilovich 2003), people consider their experiential purchases to be a better use of their money (Van Boven and Gilovich 2003; Howell and Hill 2009), and people experience greater relatedness satisfaction when reliving their life experiences (Howell and Hill 2009).

Further, some of the benefit in experiential consumption can also be explained by the hedonic cost of material consumption. One reason that materialism has been linked to decreases in well-being is that material acquisition as a strategy for seeking happiness thwarts the ability to satisfy important basic psychological needs (Kasser 2002; Ryan and Deci 2000). For example, the relationship between material buying and decreased well-being is mediated by decreased relatedness need satisfaction and vitality as well as an increased tendency to socially compare one’s purchases (Howell and Hill 2009; Howell et al. 2012). Thus, if something prevents an individual from experiencing psychological need satisfaction when spending money on life experiences, the benefit of experiential consumption will be reduced (or even eliminated). Given that it is possible for some individuals to consume life experiences for extrinsic reasons (e.g., to impress other people or for the recognition they could receive from others), it may be that the effect experiential purchases have on well-being varies by the self-determined reasons that people consume them.

1.2 Self-Determined Behaviors and Well-Being

According to SDT, when people engage in certain activities or behaviors, it is often not people’s behaviors that relate to well-being; it is the reasons for engaging in behaviors that relate to well-being (Ryan and Deci 2000). These reasons fall along a continuum of self-determination, ranging from intrinsically or autonomously motivated (e.g., engaging in behaviors because they are enjoyable, challenging, or interesting) to behaviors that are externally motivated and controlled by the environment (e.g., regulated by rewards, punishments, or threats). Intrinsically motivated behaviors lead to stable, self-determined engagement; behaviors that are externally motivated tend to be poorly maintained over time. However, SDT specifies that external motivations can be internalized to various degrees and controlled behaviors can take on properties of self-determination (e.g., introjected behavior occurs when environmental rewards and punishments are internalized, identified behavior is internally regulated and is experienced as volitional, integrated behavior aligns with other aspects of the self). However, people may also lack motivation for a behavior entirely (i.e., amotivation).

Most importantly, SDT posits that the satisfaction of basic psychological needs mediates the link between autonomous or controlled motivations and well-being (Deci and Ryan 2008). SDT assumes that three basic psychological needs—autonomy, competence, and relatedness—are critical for psychological well-being (Ryan 1995). Autonomy refers to the need to feel that one’s actions are volitional and self-determined (deCharm 1968), competence reflects the need to feel a sense of mastery and control over one’s environment (White 1963), and relatedness connotes the need to feel connected, appreciated, and loved by others (Baumeister and Leary 1995). SDT suggests that these three basic psychological needs are universally required for well-being (Deci and Ryan 2008) and that basic psychological need satisfaction is an important prerequisite for changes in well-being (Sheldon and Elliot 1999). Previous research supports these suggestions. For example, daily well-being is associated with systematic variation in basic psychological need satisfaction (Reis et al. 2000) and hourly psychological need satisfaction is correlated with momentary happiness (Howell et al. 2011).

Thus, individuals who spend money on life experiences for more autonomous reasons (e.g., because doing so aligns with their core values or because they find genuine pleasure and interest from novel, challenging activities) should experience greater well-being because their experiential consumption should satisfy basic need satisfaction; however, individuals who spend money on life experiences for more controlled reasons (e.g., to impress others) should experience less well-being because these events and activities do not fulfill basic need satisfaction. Given the likely importance of the motivations guiding experiential consumption, an understanding of individuals’ motives for experiential purchasing is imperative. However, no scale currently exists to measure to motivational structure of experiential consumption.

1.3 Previous SDT Based Motivational Scales

SDT argues that there are six factors to motivational structure (i.e., intrinsic motivation, integrated regulation, identified regulation, introjected regulation, external regulation, and amotivation). Furthermore, a number of SDT based motivational scales have revealed this six-factor structure [e.g., see scales developed by Mallett et al. (2007), Pelletier et al. (1997, 1998)]. However, the factor structure of some activities (e.g., academic motivation, sports motivation, etc.) varies. For example, the factor structure of the Situational Motivation Scale (Guay et al. 2000) revealed only four motivational styles. The Sport Motivation Scale (Pelletier et al. 1995) and the Academic Motivation Scale (Vallerand et al. 1992) share the same five-factor structure. Thus, the factor structure depends on the domain of activity. For this reason, past SDT based scales have consistently utilized the same approach: first, develop items based on the tenets of all six motivational structures, second, use exploratory factor analysis (EFA) to empirically determine the factor structure for the specific behavior, finally, employ confirmatory factor analysis (CFA) in a separate sample to verify the factor structure. Thus, we replicated the same approach—developing items based on the tenets of all six motivational structures, initiating scale development with EFA, confirming the factor structure with a CFA—that has been used to develop other SDT-based motivational scales (Situational Motivation Scale; Guayet al. 2000; Client Motivation Therapy Scale; Pelletier et al. 1997; Motivation toward the Environment Scale; Pelletier et al. 1998).

2 Overview of the Present Research

The primary aim of the present research was to develop and validate the Motivation for Experiential Buying Scale (MEBS). To do this we: (a) developed scale items and explored its factor structure; (b) confirmed its factor structure in a separate sample; (c) examined construct validity by investigating patterns of convergent validity with theoretically relevant constructs; (d) explored the temporal stability of the MEBS; and (5) demonstrated the utility of assessing motivations for experiential buying by determining the incremental validity of both a long and short version of the MEBS. Finally, we based our sample sizes (i.e., minimum sample sizes) for each of our studies on the two general sample size recommendations required to conduct factor analyses. Numerous researchers argue that the absolute number of individuals is important (see Hutcheson and Sofroniou 1999); other researchers note that the subject-to-variable ratio is used by many researchers (see Henson and Roberts 2006). We wanted to ensure our sample sizes met both thresholds. First, Hutcheson and Sofroniou recommend at least 150–300 individuals to conduct exploratory factor analyses. Second, when considering the subject-to-variable ratio, a 5:1 ratio is widely used by many researchers. We ensured that each of our samples met both of these criteria.

3 Study 1: Scale Development

The goal of Study 1 was to develop items based on the definitions of all six motivational structures and then to test the factor structure of the MEBS with exploratory factor analysis. We determined the number of factors for the MEBS using the Horn’s parallel analysis (O’Connor 2000) and tested the reliability of each factor to ensure the factors were internally consistent.

3.1 Method

3.1.1 Participants

A total of 233 students (M age = 22.20, SD = 6.56, range = 18–61; 64.4 % female; 40.3 % Caucasian) enrolled in an undergraduate psychology course at Irvine Valley Community College participated in this study. After providing informed consent, participants completed the online questionnaire, which contained items measuring motivations for buying life experiences and basic demographic information. Participants were given extra credit for their participation.

3.1.2 Item Development

To measure the six regulatory styles proposed by SDT, items were generated by the first author following the recommendations of Deci and Ryan (1985). Other SDT based motivational scales were also reviewed [e.g., the Academic Motivation Scale by Vallerand et al. (1992), the Motivation toward the Environment Scale by Pelletier et al. (1998), the Situational Motivation Scale by Guay et al. (2000), and the Revised Sport Motivation Scale by Mallett et al. (2007)]. We then deleted items that were semantically similar. A total of 43 items were selected (i.e., including 10 items measuring intrinsic motivation, eight items measuring identified regulation, six items measuring integrated regulation, six items measuring introjected regulation, seven items measuring external regulation, and six items measuring amotivation).

3.1.3 Procedure

Participants were given the definition of an experiential purchase (i.e., “an event or series of events that you personally encounter or live through” [e.g., eating out, going to a concert, traveling, etc.]) and were instructed to write about the last three times they spent their money on life experiences. After the participants described their last three experiential purchases, they read “One of the reasons I typically spend money on life experiences is…” followed by the 43 items (e.g., “because life experiences represent the kind of person I am”) and then indicated their agreement (1 = strongly agree; 7 = strongly disagree) with each item.

3.2 Results

3.2.1 The Factor Structure of the MEBS

An exploratory factor analysis (Maximum Likelihood extraction) with promax (oblique) rotation was performed on the 43 items measuring the motivational underpinnings of experiential consumption. We used a promax rotation because an oblique rotation allows the factors to correlate in order to determine a clear factor structure (see Streiner and Norman 1995). The factor analysis demonstrated that six factors had eigenvalues greater than 1.0; however, various extraction criteria suggested that a three-factor solution would be more appropriate and the factor loadings indicated that certain items could be removed. First, a three-factor solution was suggested by the scree test (Cattell 1969). Next, to confirm our three factor solution, we conducted Horn’s parallel analysis following O’Connor’s guidelines (2000) and using the Parallel Analysis Engine developed by Patil et al. (2007). Horn’s parallel analysis suggested that the first three factors were the appropriate factors to extract. Next, the pattern of factor loadings suggested that several items—including all the items measuring the introjected reasons for experiential consumption—should be dropped because they either did not load onto a factor or cross-loaded onto multiple factors. The final three-factor solution (with a total of 20 items; see Table 1) accounted for 57.25 % of the variance in these items. The variance explained by this factor structure meets the criteria recommended by Snook and Gorsuch (1989) who suggest that the factor solution should explain at least 40–50 % of the variance in the scores.

Table 1 The factor loadings of the MEBS from the exploratory factor analysis in Study 1
Full size table

The first factor (again see Table 1), an autonomous motivation for experiential consumption, consisted of 10 items (e.g., four intrinsic items [“They are part of how I’ve chosen to live my life”], two identified items [“Because I value buying life experiences”], and four integrated items [“Because I find life experiences stimulating”] all loaded onto the first factor). The second factor, a controlled motivation for experiential consumption, consisted of six items (with all six items measuring external regulation). The third factor, consisting of four items, measured no motivation (i.e., amotivation) for buying life experiences. Each of the factors demonstrated adequate internal consistency (see Table 1) as suggested by Nunnally (1978). Autonomous motivation was negatively, but not significantly, correlated with controlled motivation (r = −.12, ns); whereas, autonomous motivation was strongly and negatively correlated with amotivation (r = −.57, p < .001). Also, controlled motivation and amotivation were positively correlated (r = .28, p < .001).

3.3 Brief Discussion

Study 1 accomplished the first objective necessary to measure motivations for experiential consumption (i.e., to develop scale items based on the tenets of SDT and explore the factor structure of experiential consumption). Individuals appear to spend money on life experiences for three reasons (i.e., autonomous motivation, controlled motivation, or amotivation). However, in order to verify the factor structure of one’s motivation for experiential consumption, we recruited another sample of participants and had them complete the MEBS (using the same instructions in Study 1).

4 Study 2: Confirmatory Factor Analysis and Construct Validity

The purpose of Study 2 was to validate the proposed structure of the MEBS using confirmatory factor analysis. Also, we assessed the construct validity of the MEBS by examining the relations between autonomous motivation, controlled motivation, and amotivation for experiential consumption with theoretically related constructs (as suggested by Cronbach and Meehl 1955).

4.1 Method

4.1.1 Participants and Procedures

A total of 208 undergraduate students (M age = 26.70, SD = 8.80, range = 19–52; 59.1 % female; 60.1 % Caucasian) enrolled in a marketing course at Old Dominion University participated in this study in exchange for extra credit. After providing informed consent, participants completed the MEBS and other related measures. Specifically, participants completed: (a) the Basic Need Satisfaction Scale (Gagné 2003), which is a questionnaire measuring three psychological needs—autonomy, competence, and relatedness—theoretically necessary for optimal well-being and (b) the Subjective Vitality Scale (Ryan and Frederick 1997) which measures perceptions of energy, zeal, interests, purposes in life, and feelings of aliveness.

4.2 Results

4.2.1 Confirmatory Factor Analysis of the MEBS

Because the data met all univariate and multivariate assumptions (e.g., the data were normally distributed, including or excluding outliers did not alter the model fit, etc.), we performed a confirmatory factor analysis (CFA) using the maximum likelihood method with the AMOS 19.0 program (see Fig. 1). Results showed that the Chi-square statistic was significant, χ 2 (151, n = 199) = 287.25, p < .001. However, as noted by others (Marsh et al. 1988; McDonald and Marsh 1990), the significance of the Chi-squared statistic is highly sensitive to sample size. Specifically, Bentler and Bonett (1980) note that inferring support for a model based exclusively on the Chi-square p value is dubious. Thus, because the use and examination of many goodness-of-fit indices is widely accepted (see Hu and Bentler 1999) to determine the fit of our model, we examined various goodness-of-fit indices. Specifically, we examined: (a) the Chi-square likelihood ratio (X 2/df = 1.90), which demonstrated good fit because the value was below 3.0; (b) the Normed Fit Index (NFI), the Comparative Fit Index (CFI), and the Tucker Lewis Index (TLI) which all suggested the model fit the data well because their values (NFI = .89; CFI = .95; TLI = .93) were all above or near .90; and (c) the root mean square error of approximation (RMSEA; Steiger 1990) which demonstrated acceptable fit because the value (RMSEA = .069) was less than .08 (see Brown and Cudeck 1993). Finally, the autonomous, controlled, and amotivated latent factors predicted the appropriate items. Therefore, given that our model met the acceptable limits of several fit indices, and a CFA is a very stringent test of factor structure, these results support the three-factor structure determined from the EFA in Study 1 and we formed three motivation scores for each individual.

Fig. 1
figure 1

The confirmatory factor analysis of the MEBS in Study 2; the CFA support the 3-factor structure derived from Study 1. The Chi-square likelihood ratio and RMSEA were acceptably low (X 2/df = 1.90; RMSEA = .069). The CFI, TLI, and NFI were all acceptably high (CFI = .95; TLI = .93; NFI = .89). Finally, the autonomous, controlled, and amotivation latent factors were predicted by the appropriate items. Note modification indices suggest that several error terms needs to be correlated within factors. Error 1 is correlated with error 3 and 8. Error 2 is correlated with error 6 and 7. Error 3 is correlated with error 7. Error 4 is correlated with error 5 and 8. Error 6 is correlated with error 10. Error 9 is correlated with error 10. Error 11 is correlated with error 13 and 15. Error 12 is correlated with error 14. Error 13 is correlated with error 14

Full size image

4.2.2 The Correlations Between the MEBS and Related Measures

The means, standard deviations, and reliability coefficients for each factor and several related constructs (e.g., psychological need satisfaction, vitality, and experiential buying) are presented in Table 2. Similar to Study 1, autonomous motivation was not significantly correlated with controlled motivation (r = −.02, ns); whereas, autonomous motivation was moderately and negatively correlated with amotivation (r = −.34, p < .001). Also, controlled motivation and amotivation were moderately and positively correlated (r = .39, p < .001).

Table 2 Means, standard deviations, reliability coefficients, and correlations between the MEBS and related measures from Study 2
Full size table

4.2.3 Convergent Validity

Individuals who spend money on life experiences for autonomous reasons report moderately higher overall psychological need satisfaction, autonomy, competence, relatedness, and vitality. In contrast, individuals who spend money on life experiences for controlled reasons and without reason (i.e., amotivation) report moderately lower overall psychological need satisfaction, autonomy, competence, and relatedness.

4.2.4 Incremental Validity

We tested for the incremental validity by regressing overall psychological need satisfaction, autonomy, relatedness, competence and vitality onto the three extracted factors (see Table 3). As expected, increased autonomous motivation predicted a moderate increase in overall psychological need satisfaction, autonomy, competence, relatedness, and vitality even when controlling for controlled motivation and amotivation. Also, increased controlled motivation predicted a moderate decrease in overall psychological need satisfaction, autonomy, competence, and relatedness even when controlling for autonomous motivation and amotivation. Interestingly, amotivation did not explain unique variance in psychological need satisfaction nor vitality when controlling for autonomous motivation and controlled motivation.

Table 3 Multiple regressions predicting overall psychological needs, autonomy, competence, relatedness and vitality in Study 2
Full size table

4.3 Brief Discussion

The results confirmed the factor structure of the MEBS in a separate sample using CFA, thereby, accomplished the objective of Study 2. Specifically, the CFA demonstrated that the autonomous, controlled, and amotivated latent factors predicted the appropriate items. Also, the regression analyses demonstrated that autonomous and controlled motivation for experiential consumption explained unique variance in psychological need satisfaction and vitality. However, these validity tests correlated self-reports from the MEBS with self-reports from other well-being measures. Thus, in Study 3 we assessed the convergent validity of the MEBS by correlating participants’ responses on the MEBS with judges’ ratings of the intrinsic and extrinsic motivation from the participants’ open-ended descriptions of their past three experiential purchases.

5 Study 3: Judges’ Ratings of Motivation

The central aim of the third study was to provide additional support for the construct validity of the MEBS through judges’ ratings of participants’ motivations. In this study, the instructions for the MEBS were identical to all previous administrations with the addition that we informed the participant they needed to also describe “why [they made] these purchases.” Then, we recruited five undergraduate students enrolled in a psychology course from San Francisco State University, who were blind to everything but the purchase descriptions, to rate the degree to which the participants’ descriptions appeared to be intrinsically and extrinsically motivated.

5.1 Method

5.1.1 Participants and Procedures

A total of 139 participants from multiple psychology courses at San Francisco State University (M age = 23.78, SD = 9.10, range = 18–67; 84.8 % female; 37.9 % Caucasian) completed an online survey. Participants completed: (a) the MEBS, (b) the Basic Need Satisfaction Scale (Gagné 2003), and (c) the Subjective Vitality Scale (Ryan and Frederick 1997).

5.1.2 Coding

First, we provided the judges with the definitions of intrinsic and extrinsic motivations. Next, coders rated their level of agreement with the following statements (1 = strongly disagree; 5 = strongly agree): (1) these purchases were made for intrinsic reasons and (2) these purchases were made for extrinsic reasons. If the majority of the coders could not rate the level of motivation based on the text, the rating for that subject was excluded from all analyses. Eight participants were excluded as a result. The reliability between the judges was adequate as demonstrated by the intra-class correlations for intrinsic motivation (ICC = .66) and extrinsic motivation (ICC = .61).

5.2 Results

5.2.1 Judges and Self Ratings: Inter-Correlations and External Correlations

As seen in Table 4, three of the important convergent validity correlations (between the judges’ ratings of perceived motivation for experiential buying and the participants’ self-ratings on the MEBS) were moderately associated and significant. Overall, when judges rated the purchases as more intrinsically motivated, the participants also rated themselves higher in autonomous motivation (r = .26, p < .01) and participants reported lower controlled motivation (r = −.18, p < .05). When judges rated the purchases as more extrinsically motivated participants rated themselves higher in amotivation (r = .19, p < .05). Interestingly, the judges’ ratings of extrinsic motivation and the participant’s self-reported controlled motivation did not correlate significantly. We suspected that this might be due to variations in the description length. Consequently, we conducted a follow-up analysis that examined only those individuals who wrote description of more than 260 characters (N = 96). For this subsample, the judges’ ratings of extrinsic motivation and participants’ self-reported controlled motivation were positively correlated (r = .17, p < .10). Furthermore, all previous significant correlations remained significant. Next, we correlated judges’ ratings with the self-reported psychological need satisfaction and vitality (again see Table 4). When the judges rated the purchases as more intrinsically motivated, the participants also reported more psychological need satisfaction (r = .17, p < .10), relatedness (r = .23, p < .05) and vitality (r = .31, p < .01).

Table 4 Self-judges convergent validity correlations in Study 3
Full size table

5.3 Brief Discussion

The results of Study 3 are consistent with the results from study 2 and further confirmed the construct validity of the MEBS. Specifically, self-reported autonomous motivation for experiential consumption was related to increase self-report of psychological need satisfaction and judges’ ratings of intrinsic motivation. The next goal in developing the MEBS was to examine the temporal stability of the MEBS.

6 Study 4: Test–retest Reliability and Continued Validation

The first goal of Study 4 was to examine the test–retest reliability of the MEBS. The second goal was to further examine the construct validity of the MEBS by correlating it with psychological need satisfaction and additional well-being measures.

6.1 Method

6.1.1 Participants and Procedures

A sample of 92 undergraduate students (M age = 24.67, SD = 6.97, range = 19–54; 73.9 % female; 43.5 % Caucasian) from San Francisco State University (SFSU) completed the MEBS survey on one occasion and then again 14 days later. The question order was randomized for each participant across both administrations of the MEBS.

In addition to the MEBS, participants completed: (a) the Basic Need Satisfaction Scale (Gagné 2003); (b) the Subjective Vitality Scale (Ryan and Frederick 1997); (c) the Revised UCLA Loneliness Scale (R-UCLA; Russell et al. 1980), which measures participants’ subjective experience of isolation; and (d) the Flourishing Scale (FS; Diener et al. 2010), which measures important aspects of human functioning (e.g., the presence of positive interpersonal relationships, meaning, and purpose in life).

6.2 Results

6.2.1 Internal Consistency and Temporal Stability of the MEBS

For time 1 and 2, respectively, the reliability coefficients of the autonomous motivation (α = .84, .89), controlled motivation (α = .84, .84), and amotivation (α = .79, 85) scales demonstrated good internal consistency (see Table 5). Also, the MEBS demonstrated adequate temporal stability. The test–retest Pearson correlations were .60 for autonomous motivation, .68 for controlled motivation, and .64 for amotivation (all p’s < .001). While researchers generally recommend test–retest correlations of .70 or greater, our results are consistent with other SDT-based scales which report similar test–retest correlations (see Pelletier et al. 1998).

Table 5 Means, SD, alpha reliabilities and test–retest reliability of the MEBS in Study 4
Full size table

6.2.2 Convergent Validation of the MEBS

We tested the convergent validity of the MEBS by correlating each motivational subscale with various measures of psychological need satisfaction and well-being at both time points (see Table 6 for all correlational results reported below). First, the inter-correlations between the motivational scales were similar to the relations found in Study 1, Study 2, and Study 3. That is, at both time 1 and time 2, autonomous motivation was not correlated with controlled motivation (r = .01 and .10, ns) and autonomous motivation was moderately and negatively correlated with amotivation (r = −.33, p < .01; r = −.27, p < .05), whereas, controlled motivation was moderately and positively correlated with amotivation (r = .53, p < .001, r = .37, p < .01). Second, at both time points, individuals who spend money on life experiences for autonomous reasons reported more overall psychological need satisfaction, autonomy, competence, relatedness, vitality, and flourishing as well as less loneliness (see Table 6 again). On the other hand, individuals spend money on life experiences for controlled reasons reported, at both time points, less overall psychological need satisfaction and autonomy. Finally, those who did not know why they spent money on life experiences reported, at both time points, less overall psychological need satisfaction and more loneliness.

Table 6 Correlations between autonomous, controlled and amotivation with outcome measures in Study 4
Full size table

6.3 Brief Discussion

The results of Study 4 support the previous convergent validity results and demonstrate that the MEBS exhibits acceptable temporal reliability. The next goal was to test the predictive validity of the MEBS above and beyond people’s general tendency to spend money on life experiences over material possessions (Howell et al. 2012).

7 Study 5: Testing the Incremental Validity and the MEBS

A recent finding on experiential consumption has demonstrated that a preference for spending money on life experiences, as opposed to material items, is related to increased psychological need satisfaction (Howell et al. 2012). As a result, we aimed to test the incremental validity of the MEBS above and beyond the preference for experiential buying in predicting psychological need satisfaction in study 5. Also, as shorter scales “eliminates item redundancy and therefore reduces the fatigue, frustration, and boredom associated with answering highly similar questions repeatedly” (Robins et al. 2001, p. 152), in addition to testing the incremental validity of the MEBS, we aimed to create an equivalent 9-item (i.e., three items per motivation) version of the MEBS.

7.1 Method

7.1.1 Participants

We recruited a sample of 241 undergraduate students (M age = 23.49, SD = 6.03, range = 18–71; 78.3 % female; 39.8 % Caucasian) from SFSU to complete the MEBS and the Experiential Buying Tendency Scale (EBTS; Howell et al. 2012). In addition to the MEBS and the EBTS, participants completed: (a) the Basic Need Satisfaction Scale (Gagné 2003) and (b) the Subjective Vitality Scale (Ryan and Frederick 1997).

7.1.2 Procedures

We selected three MEBS items from each of the MEBS factors based on several criteria from previous studies that developed brief scales of individual difference constructs (Gosling et al. 2003; Rammstedt and John 2007; Saucier 1994). First, we attempted to maximize breadth of coverage by selecting the three items that best represented each motivation. Second, we selected items that were not deemed to be evaluatively extreme. Third, we selected items based on the simple-structure pattern of their factor loadings in factor analyses of all 20 items. The final nine items are described at the bottom of Appendix.

7.2 Results

7.2.1 Convergent and Discriminant Validity

The means, standard deviations, alpha coefficients, and convergent correlations between the MEBS-9 and MEBS-20 are displayed in bold face in Table 7. The alpha coefficients for the 9-item version were weaker (although all above .70) than the long version; however, this was expected because of the reduced number of items. However, the mean convergent correlation of .95 is much larger than the absolute mean of the off-diagonal discriminant correlation of .20; this suggests that while the 20-item inventory tends to be more reliable, the nine-item short version may be an adequate substitute. Also supporting the validity of the MEBS-9, the correlations between the scales were similar when using the 9-item and 20-item versions (again see Table 7). That is, autonomous motivation was not correlated with controlled motivation (r = .02, ns for the 20-item version; r = −.05, ns for the 9-item version) and was moderately and negatively correlated with amotivation (r = −.22, p < .001 for the 20-item version; r = −.20, p < .001 for the 9-item version). Lastly, controlled motivation and amotivation were positively correlated (r = .40, p < .001 for the 20-item version; r = .31, p < .001 for the 9-item version; Table 7).

Table 7 Convergent correlations between the short MEBS and long MEBS in Study 5
Full size table
Table 8 Hierarchical regressions predicting incremental validity on psychological needs, autonomy, competence, relatedness and vitality, over EBTS using the MEBS-20 in Study 5
Full size table

7.2.2 The Incremental Validity of the MEBS-20 and MEBS-9

Because the tendency to make experiential purchases is associated with an increase in psychological need satisfaction (Howell et al. 2012), we used hierarchical regression analyses to explore the relative contributions of the MEBS-20 and the MEBS-9 beyond the general tendency toward experiential buying (as measured with the EBTS). That is, we entered the EBTS in step 1 of all regression analyses and then the MEBS-20 or the MEBS-9 in step 2; see Tables 8 and 9). First, in all of the regression analyses, even though a preference for experiential buying was associated with increased psychological need satisfaction and vitality, one’s motivation for experiential consumption explained unique variance in all five outcomes. Notably, in each model, autonomous motivation for experiential consumption was a significant predictor of increased psychological need satisfaction (including each facet of psychological well-being) and vitality. Also, controlled motivation for experiential consumption was a significant predictor of decreased psychological need satisfaction (specifically, decreased autonomy and competence). When we used the MEBS-9 to predict psychological need satisfaction and vitality, we replicated these patterns (see Table 9). Specifically, in all of the regression analyses, the MEBS-9, explained unique variance in all five outcomes, autonomous motivation for experiential consumption was a significant predictor of increased psychological need satisfaction and vitality, and controlled motivation for experiential consumption was a significant predictor of decreased psychological need satisfaction (specifically, decreased autonomy and competence).

Table 9 Hierarchical regressions predicting incremental validity on psychological needs, autonomy, competence, relatedness and vitality, over EBTS using the MEBS-9 in Study 5
Full size table

8 General Discussion

The value of experiential pursuits has been promoted since Fromm (1976), more recently by academic scholars (Caprariello and Reis 2010; Carter and Gilovich 2010; Howell and Hill 2009), and propagated by social media (Landau 2009; Rettner 2010). The literature on experiential consumption has consistently demonstrated that experiential purchases, compared to material purchases, makes people happier (Howell et al. 2012; Nicolao et al. 2009; Van Boven and Gilovich 2003) and that the fulfillment of psychological needs accounts for much of this relationship (Howell and Hill 2009). To our knowledge, the motivational antecedents of experiential consumption have not been explored. To address this gap, we developed and validated the MEBS—a measure to assess autonomous and controlled motivations for experiential consumption.

Across five studies we demonstrated that individuals who buy life experiences for autonomous reasons reported more autonomy, competence, relatedness, flourishing, vitality, and less loneliness. On the other hand, individuals who buy life experiences for controlled reasons reported less autonomy and competence, even after controlling for their general buying tendencies and their autonomous motivation. Thus, our results highlighted a scenario when experiential consumption may not increase psychological need satisfaction—when life experiences are consumed for controlled reasons. Specifically, our results suggest that the benefits of the experiential recommendation may be amplified when consumed for autonomous reasons or may be attenuated when consumed for controlled reasons.

8.1 The Motivational Structure of Experiential Consumption

We demonstrated the necessity of measuring the motivational antecedents of experiential buying as well as the benefit of applying SDT towards understanding the difference between the content (a preference for spending money on life experiences) and motives (autonomous vs. controlled spending on life experiences) of human behaviors (see Vansteenkiste et al. 2010 for a review). Curiously, even though SDT predicts six regulatory styles, and the items we wrote for the MEBS utilized the principles of SDT, the factor analyses of our data suggested that the motivational structure of experience consumption can be accounted for by autonomous, controlled, and amotivation. That is, because intrinsic, integrated, and identified items loaded onto one overall autonomous motivation factor, introjected items did not load onto any factor, and controlled and amotivation items formed their own factors, we believe, based on our data, a three-factor structure may be the appropriate structure for understanding the motivational antecedents of experiential buying. Also, in general, self-report measures can suffer from several biases, including the desire to respond in a socially desirable manner (Paulhus and Vazire 2007). Therefore, the result from Study 3—where judges’ and participants’ ratings of the motivational antecedent of previous consumption was correlated—minimizes social desirability biases concerns. Thus, our scale deviated from the predictions SDT because we allowed the data to drive the development of the scale, an approach consistent with prior research (Pelletier et al. 1995; Vallerand et al. 1992).

It is worthy to consider why items that measure introjected regulation failed to load reliably on any factors. Introjected regulation tends to focus on engaging in behaviors to maintain or enhance self-worth (Deci and Ryan 2004). However, experiential purchases, more so than material purchases, tend to be closely associated with the “true, essential self” (Carter and Gilovich 2012). If this is true, then spending money on life experiences to improve or maintain self-esteem may be a less central concern than other forms of external regulation in experiential consumption. Thus, introjected items would not be expected to load onto either autonomous or controlled motivations for experiential consumption. Furthermore, existing conceptualizations of introjected motivation tap largely into avoidance aspects of introjections (e.g., avoiding shame or guilt), rather than approach aspects (e.g., seeking self-worth; see Assor et al. 2009).Footnote 2 Given that experiential consumption is marked more strongly by reward-seeking and approach motivation than by avoidance motivation (Howell et al. 2012), it is possible that our items, by conforming to traditional measures of introjected regulation, focused on avoiding negative outcomes. Future research should attempt to measure introjected regulation (i.e., behaviors to maintain or enhance self-worth) of experiential consumption with approach-orientated items as well as avoidant-orientated items.

8.2 Limitations and Future Directions

Nevertheless, a few limitations are present. For example, because of the correlational nature of the data we cannot demonstrate causality. Though this is universal to all survey development, for the MEBS, this limitation is especially important in understanding how motives impact well-being because individuals’ motivation can be affected by their general feeling of psychological need satisfaction (Sheldon 2011; Sheldon and Gunz 2009; Sheldon et al. 2010; see Deci et al. 1999 for discussion of causality issues in motivational research). In fact, controlled motivations for experiential buying may be a direct result of thwarted need satisfaction, whereas autonomous motivations may be a direct result of satisfied needs. The two constructs are deeply and strongly intertwined. When a person’s psychological needs are satisfied, the regulation of experiential buying behaviors may be more internalized and well-being outcomes more likely to be maximized. When a person’s psychological needs are thwarted the regulation of experiential buying behaviors may be more controlled and well-being outcomes less likely to be maximized. Determining how motives for experiential consumption develop would be instrumental for future studies of experiential purchasing.

It is important to note that researchers have debated how to conceptualize, and measure well-being (see Busseri and Sadava 2011 for a review). Throughout this research, we measured well-being in ways consistent with SDT. That is, we used measures that tapped into components of well-being that tend to go beyond hedonic pleasure and into a sense of fulfillment or eudaimonic well-being. Thus, we encourage future studies to replicate our results with more pleasure oriented measure of well-being (e.g., subjective happiness or positive and negative affect). However, it is likely that individuals who have their psychological needs satisfied will also report high levels of hedonic well-being. Consequently, we would expect autonomously motivated experiential consumers to experience more hedonic well-being than controlled and amotivated experiential consumers.

9 Conclusion

Recently it was suggested that money spent with the intention of satisfying psychological needs may lead to increased well-being (Howell and Howell 2008). Further, studies have now confirmed that spending money on life experiences, rather than buying material items, results in greater experienced SWB (Howell and Hill 2009; Nicolao et al. 2009; Van Boven and Gilovich 2003). Our studies represent an effort to understand when experience consumption may not benefit the individual. Consistent with SDT, and replicating Howell et al. (2012) we demonstrated that habitual experiential consumption is associated with increased well-being—however, the benefits of experiential buying can be enhanced when pursued for autonomous reasons and eliminated when pursued for controlled reasons. Now that the benefits of experiential buying are becoming clearer, it is our hope that the development of the MEBS will stimulate future research to investigate the contexts in which experiential purchases do, and do not maximize utility. It appears that the experiential recommendation should not only encourage individuals to allocate their resources toward doing rather than having (Van Boven and Gilovich 2003; Fromm 1976), but more importantly, to do it with self-determined reasons.