When Knowledge Is Demotivating

Abstract

People find it appealing to have more options to choose from, but the provision of choice often leads to adverse consequences for decision makers’ motivation, satisfaction, and willingness to act. We propose that the effect of the number of choice options on willingness to purchase is moderated by people’s subjective knowledge (SK). The results of three studies provide converging evidence that, paradoxically, people who feel unknowledgeable (low-SK people) in a certain domain are especially willing to purchase when more choice options are available, which is consistent with the notion of “more is better.” This pattern is reversed for people who feel knowledgeable (high-SK people), which is consistent with prior evidence for choice overload. We also show that this pattern is influenced by the informativeness of the features of the available choice options and that subjective knowledge mediates this effect.

Keywords

subjective knowledge choice-set size choice overload

Choice architects, such as policymakers, marketers, and financial advisors, have long debated how many choice options to offer decision makers. Previous research has found mixed results (Scheibehenne, Greifeneder, & Todd, 2010). On the one hand, some findings indicate a more-is-better effect: People find larger choice sets more appealing, more intrinsically motivating, and more enjoyable to choose from (Botti & Iyengar, 2004; Iyengar & Lepper, 2000). Larger choice sets also allow decision makers to learn more about the range of available options (Bellenger & Korgaonkar, 1980), match the tendency to seek variety (Kahn, 1995; McAlister, 1982), and increase the chances that decision makers will find a good match for their preferences (Arrow, 1959).

On the other hand, other research suggests that choosing from large choice sets overloads decision makers. In their seminal article, Iyengar and Lepper (2000) showed that choice overload results in lower willingness to purchase any option because decision makers experience more difficulty and frustration during the decision-making process. More recently, Chernev (2003a, 2003b) demonstrated that choosing from a large choice set leads to lower confidence in choice. Finally, choice overload has also been associated with postdecisional regret and dissatisfaction with choice (Diehl & Poynor, 2010; Mogilner, Rudnick, & Iyengar, 2008).

In their report on a meta-analysis that reexamined these conflicting results, Scheibehenne et al. (2010) proposed that moderating variables may account for both the positive and negative consequences of large choice sets. Drawing on recent findings that choice is driven primarily by decision makers’ subjective knowledge, we offer the counterintuitive hypothesis that subjective knowledge moderates choice overload such that decision makers who feel knowledgeable in the choice domain experience choice overload, whereas decision makers who feel unknowledgeable benefit from the provision of more choice options.

Previous research has distinguished expertise, or actual knowledge, from subjective knowledge—decision makers’ beliefs about their state of knowledge (Alba & Hutchinson, 1987, 2000; Hadar, Sood, & Fox, 2013). For example, recognizing that red wine can be made of six different grapes is an example of actual knowledge, whereas the belief that one knows a lot about red wines is an example of subjective knowledge. Research has found that people use their subjective knowledge to strategically guide future behavior, often independently of their actual knowledge (Fox & Weber, 2002; Hadar et al., 2013). For example, people with higher subjective knowledge about a product express less interest in searching for product-related information because they believe new information will be redundant with their current knowledge (Urbany, Dickson, & Wilkie, 1989; Wood & Lynch, 2002). In contrast, people with lower subjective knowledge seek out product information more extensively and view new product information as more important (Park, Gardner, & Thukral, 1988; Urbany et al., 1989).

Subjective knowledge has also been found to correlate positively with decision confidence and with willingness to act. Analyses controlling for actual knowledge have shown that people with higher subjective knowledge are more confident in their ability to make effective decisions (Park, Mothersbaugh, & Feick, 1994; Parker, Bruine de Bruin, Yoong, & Willis, 2012; Radecki & Jaccard, 1995) and are consequently more willing to act on their beliefs (Bearden, Hardesty, & Rose, 2001; Fernandes, Lynch, & Netemeyer, 2014; Raju, Lonial, & Mangold, 1995). For example, we (Hadar et al., 2013) previously reported that decision makers were more likely to invest in options they felt more knowledgeable about, even when they had less actual knowledge about these options, because they felt more confident that these were better investments.

In the experiments reported here, we compared decision makers’ willingness to purchase from large choice sets and from small choice sets. Drawing on previous research, we hypothesized that the effect of choice-set size on willingness to purchase depends largely on people’s subjective knowledge of the options’ domain. We predicted that people who had high subjective knowledge in a given domain (high-SK people) would be less likely to purchase from large choice sets than from small choice sets, whereas people who had low subjective knowledge in a given domain (low-SK people) would be more likely to purchase from large choice sets than from small choice sets.

Subjective Knowledge and Choice-Set Size

Choice sets have been associated with two characteristics: First, large choice sets are more difficult and frustrating to choose from than small choice sets (Iyengar & Lepper, 2000). Second, if large choice sets include unique options, they convey more information about the choice domain (e.g., the range of options and types of option features) than do small choice sets (Bellenger & Korgaonkar, 1980). We propose that these characteristics affect low- and high-subjective knowledge decision makers’ willingness to purchase differently.

Large choice sets expose high-SK decision makers to an excessive amount of information (Urbany et al., 1989; Wood & Lynch, 2002) and lead to frustration because such decision makers lack interest in receiving more information (Iyengar & Lepper, 2000). In addition, high-SK decision makers expect choosing to be easy because they feel confident about their knowledge and ability to make a good choice (regardless of the number of options; Park et al., 1994; Parker et al., 2012; Radecki & Jaccard, 1995). Because of such expectations, they often find choosing from large choice sets (but not from small choice sets) more difficult than initially expected (Lichtenstein, Fischhoff, & Phillips, 1982). Consequently, high-SK people may feel less knowledgeable and less confident in their ability to make a good choice when there are more options to choose from (Schwarz, 2004). Thus, a large choice set lowers their subjective knowledge and reduces their willingness to purchase.

Conversely, low-SK people expect choosing to be difficult. At the same time, choosing from large choice sets exposes them to the additional product information that they desire (Park et al., 1988). Being exposed to more information may make low-SK people feel more knowledgeable and more confident in their ability to make a good choice when choosing. As a result, a large choice set increases their subjective knowledge and their willingness to purchase.

In summary, we hypothesize that subjective knowledge (independently of actual knowledge) moderates the relationship between choice-set size and willingness to purchase. High-SK people are less willing to purchase when they are choosing from a large choice set than when they are choosing from a small set, whereas low-SK people are more willing to purchase when they are choosing from a large choice set than when they are choosing from a small set. Furthermore, we propose that this effect is mediated by the effect of choice-set size on subjective knowledge. Figure 1 depicts our theoretical model. We next report the results of three studies that formally tested our hypothesis.

Fig. 1.

The proposed model. Large sets have different effects on people with high and low subjective knowledge (high-SK people and low-SK people). High-SK people find large sets to be demotivating compared with small choice sets, because the large sets increase choice difficulty and are frustrating to choose from because they offer unwanted product information. Low-SK people find large sets more appealing than small choice sets because they provide more information, and such people are not deterred by the difficulty of choosing because they expect choice to be difficult. Therefore, large sets, relative to small sets, reduce high-SK people’s willingness to purchase (WTP) but enhance low-SK people’s WTP.

Study 1

In Study 1, we sought to establish the moderating effect of subjective knowledge on the relationship between choice-set size and willingness to purchase. Subjective knowledge was manipulated by varying participants’ comparative ignorance. Research shows that having some decision-makers compare themselves with a more knowledgeable person, and others compare themselves with a less knowledgeable person, differentially influences their subjective knowledge of the domain in question while holding actual knowledge and experience constant (Fox & Weber, 2002; Hadar et al., 2013; See, 2009).

Method

A sample of 300 adults was recruited through Amazon.com’s Mechanical Turk (MTurk). The results of 284 participants (52.1% female, 47.9% male; mean age = 35.6 years, SD = 12.7) who completed the survey were included in our analyses. We wanted at least 50 subjects per condition, and we assumed that some participants would not complete the survey, so we decided to recruit 300 people. Participants were randomly assigned to one of four conditions in a 2 (subjective knowledge: low or high) × 2 (choice set: large or small) design.

Participants were first informed that they would be asked to choose one coffee from a menu. Next, participants in the low-SK condition were informed that the survey was being presented to a sample of MTurk users and professional coffee tasters (a more knowledgeable group), whereas participants in the high-SK condition were informed that the survey was being presented to a sample of MTurk users and high-school students (a less knowledgeable group). Participants then rated their subjective knowledge of coffee relative to that other group’s knowledge (see Table 1).

Table 1.

Manipulation-Check Items

Manipulation	Item	Rating scale	Study
Relative subjective knowledge	Rate your knowledge of [product] compared to [comparison group].	1 = much less, 4 = about the same, 7 = much more	1, 3
Rated absolute subjective knowledge	How knowledgeable do you feel about [product]?	1 = I know very little, 7 = I know a lot	2, 3
Rated absolute subjective knowledge	How confident are you that you can make a good choice of red wine?	1 = not at all, 7 = very much	2
	Rate your knowledge of red wines compared to the average consumer.	1 = much less, 4 = average, 7 = much more	2
	Rate your confidence in using red wine information compared to the average consumer.	1 = much less, 4 = average, 7 = much more	2
	How confident are you that you made a good choice of video game?	1 = not at all, 7 = very much	3
Choice-set size	How much choice do you feel you were offered in terms of the [product] selection?	1 = no choice at all, 7 = a lot of choice	1, 2, 3
	When initially given the task to pick one [product] from the list, do you think the selection should have included more options to choose from?	1 = I had too few options to choose from, 7 = I had too many options to choose from	1, 2, 3
Perceived quality	How would you rate the overall quality of the red wine options you were presented with?	1 = low quality, 7 = high quality	2
	Overall, how positively do you feel about the wine options you were presented with?	1 = not at all positive, 7 = very positive	2
Attribute informativeness	To what extent the information you received about the different red wines was informative in making the wine choice?	1 = not at all, 7 = very much	2
	To what extent the information you received about the different red wines was useful in making the wine choice?	1 = not at all, 7 = very much	2
	To what extent the information you received about the different red wines was sufficient in making the wine choice?	1 = not at all, 7 = very much	2

Next, participants chose 1 coffee from a menu of 5 or 25 options, all of which were reported to be in the price range of U.S. $13 to $15 per lb. The coffee options presented in the small choice set were rotated such that every option encountered in the large choice set was encountered in the small set as well; there were no identical options (replicates) in either set. The coffee options were described by two features: type (e.g., espresso forte) and flavor (e.g., tangy), adapted from Mogilner et al. (2008).

The dependent measure of willingness to purchase was adapted from Iyengar and Lepper (2000, Study 3). After choosing one coffee from the menu, participants were told that 2 participants would be randomly selected to receive additional compensation—either 1 lb of the coffee they had chosen or $5. Participants indicated which of these they would prefer if they were selected. It was noted that they would not be able to change their preference later. Our dependent measure was the percentage of participants who chose the coffee (i.e., who were willing to “purchase” the coffee rather than receive the $5 they could have had otherwise). Finally, participants filled out two items as a check for choice-set size manipulation (see Table 1; α = .70).

Results and discussion

Manipulation checks

As expected, participants’ subjective knowledge was lower when they rated their knowledge of coffee compared with that of expert coffee tasters (M = 2.41) than when they compared their knowledge with that of high-school students (M = 5.52), t(282) = 20.9, p < .0001. Both means were significantly different from the midpoint of the scale, labeled about the same (ps < .0001). In addition, the small choice set was perceived as smaller than the large set (M = 4.31 vs. M = 5.48), t(282) = 9.5, p < .0001.

Choice

A logistic regression revealed the predicted SK Condition × Choice-Set Size interaction, χ²(1, N = 284) = 7.5, p = .006: Low-SK participants were more likely to choose the coffee over the $5 when the choice set was large (39%) than when it was small (24%), χ²(1, N = 143) = 3.7, p = .05. In contrast, high-SK participants were less likely to choose the coffee over the $5 when the choice set was large (11%) than when it was small (24%), χ²(1, N = 141) = 4.1, p = .04. The main effect of choice-set size was nonsignificant, χ²(1, N = 284) = 0.14, p = .71. Finally, overall, low-SK participants (31%) were more likely than high-SK participants (18%) to choose coffee over the $5, χ²(1, N = 284) = 7.2, p = .007.

The results of Study 1 supported our hypothesis that subjective knowledge moderates willingness to purchase from large and small choice sets. High-SK participants were less likely to purchase when choosing from many options than when choosing from few options, whereas low-SK participants were more likely to purchase when choosing many options than when choosing from few options.

We propose that this effect occurs because of the tendency of larger choice sets to be more difficult and frustrating to choose from, as well as to convey more information about the choice options. We tested this hypothesis in Study 2. In Study 1, the features provided for each option were informative for selecting a coffee; by implication, the larger choice sets were more informative than the smaller choice sets. In Study 2, we varied the informativeness of the features provided. We predicted that if the larger choice set did not convey relevant information, choosing from it should not be any more difficult than choosing from the small choice set (because the choice might be more arbitrary), and such a noninformative choice set might not affect decision makers’ subjective knowledge. Both high- and low-SK people should thus be equally likely to purchase from large and small choice sets when the information conveyed about the options is noninformative.

Study 2

In Study 2, our goal was to replicate Study 1 using another set of options and a different manipulation of subjective knowledge. We also tested the interaction among the informativeness of the features provided, SK condition, and choice-set size. In addition, we contrasted the effects of subjective knowledge and actual knowledge, and we checked to be sure that the perceived quality of the options did not differ between the choice sets.

Method

A sample of 800 MTurk users was recruited to participate in this study. The results of 719 participants (65% female, 35% male; mean age = 34.4 years, SD = 11.8) who completed the survey were included in our analyses. We wanted at least 100 subjects per condition but did not reach that target because some participants did not complete the survey. Participants were randomly assigned to one of eight conditions in a 2 (subjective knowledge: low or high) × 2 (choice set: large or small) × 2 (feature informativeness: low or high) design.

The subjective-knowledge manipulation, adopted from Moorman, Diehl, Brinberg, and Kidwell (2004), included a measure of actual knowledge, which enabled us to estimate the relative influence of subjective knowledge and actual knowledge on willingness to purchase. Participants filled out a survey of their actual knowledge of red wines and were then provided false feedback regarding their performance. They were informed that their knowledge scores were compared with those of 1,000 MTurk users; participants in the low-SK condition were told that their scores were in the 10th percentile, and those in the high-SK condition were told that their scores were in the 90th percentile. Participants then rated their subjective knowledge (four items; see Table 1; α = .94).

Next, participants chose a red wine from a set of 5 or 25 options, all reported to be priced at about $10. Two features that were either informative or noninformative, depending on condition, were used to describe each wine.¹ After choosing a wine, participants were told that 2 participants would be randomly selected to receive additional compensation—either a bottle of the wine they had chosen or $5. Participants indicated which of these they would prefer if they were selected. It was noted that they would not be able to change their preference later. Finally, participants filled out manipulation-check items (see Table 1) regarding the choice-set size (α = .69), perceived quality of the wine offerings (α = .80), and feature informativeness (α = .93). Before the survey terminated, participants were debriefed on the survey goals and the use and need for the false-feedback manipulation. They were then given their true actual knowledge scores.

Results and discussion

Manipulation checks

As expected, wine-knowledge ratings from participants in the low-SK condition (M = 3.07) were lower than those of participants in the high-SK condition (M = 4.52), t(717) = 16.8, p < .0001. Informativeness ratings of participants in the low-feature-informativeness condition (M = 2.91) were lower than those of participants in the high-feature-informativeness condition (M = 4.57), t(715) = −16.3, p < .0001. In addition, the small choice set (M = 4.22) was perceived as smaller than the large choice set (M = 4.98), t(715) = 10.4, p < .0001, although the selection of wine had similar perceived quality in the small-set-size condition (M = 4.82) and in the large-set-size condition (M = 4.88), t(715) = 0.8, p = .42.

Choice

We performed a logistic regression in which the likelihood of preferring the chosen wine over the dollar amount was regressed on feature informativeness, SK condition, choice-set size, and the interactions among them. To control for the possible impact of actual knowledge, we included actual knowledge as a covariate. The results revealed a marginally significant three-way interaction among SK condition, choice-set size, and feature informativeness, χ²(1, N = 719) = 3.33, p = .068. Further analyses revealed that the SK Condition × Choice-Set Size interaction found in Study 1 was replicated when feature informativeness was high, χ²(1, N = 360) = 8.12, p = .004: Low-SK participants were more likely to select the wine than the $5 when the choice set was large (34%) than when it was small (22%; Fig. 2), χ²(1, N = 180) = 3.3, p = .069. In contrast, high-SK participants were less likely to select the wine when the choice set was large (24%) than when it was small (38%), χ²(1, N = 180) = 3.9, p = .031. However, no interaction emerged when feature informativeness was low, χ²(1, N = 359) = 0.03, p = .87. Actual knowledge was not a significant predictor of choice, χ²(1, N = 719) = 1.57, p = .21. There was a marginally significant main effect of feature informativeness, such that participants in the low-feature-informativeness condition were more likely to select the wine than the $5 (24%) than participants in the high-feature-informativeness condition (30%), χ²(1, N = 719) = 3.36, p = .07. However, no main effect of SK condition, χ²(1, N = 719) = 0.1, p = .75, or of choice-set size, χ²(1, N = 719) = 0.01, p = .99, was found.

Fig. 2.

Results from Study 2. The percentage of participants who chose wine rather than cash is shown as a function of feature informativeness and subjective knowledge (SK) for both set-size conditions.

The results of Study 2 replicated those of Study 1, showing an interaction between SK condition and choice-set size when we used another set of options and a different type of subjective knowledge manipulation. In addition, Study 2 revealed that this interaction occurs only to the extent that the information conveyed in the choice set is informative. The results also rule out the possibility that the SK Condition × Choice-Set Size interaction is driven by differences in the perceived quality of the options in the choice sets or by actual knowledge.

Study 3

In Study 3, we directly tested postchoice self-rated subjective knowledge, which was proposed to underlie the SK condition × Choice-Set Size interaction. We hypothesized that large choice sets would frustrate high-SK people and undermine their confidence in their knowledge, so that post-choice self-rated subjective knowledge would be lower, leading to lower willingness to purchase in the large-set-size condition than in the small-set-size condition. However, we hypothesized that large choice sets would make low-SK people feel more knowledgeable, so that post-choice self-rated subjective knowledge would be higher, leading to higher willingness to purchase in the large-set-size condition than in the small-set-size condition.

Method

We recruited 145 undergraduate students (62.8% female, 37.2% male) to participate in this paper-and-pencil study. Sample size was determined by the maximum number of participants we were able to recruit for the data-collection session.

Participants were informed that they would be asked to choose a video game. As in Study 1, participants were next informed that the survey was being presented to a sample of undergraduate students (i.e., the participants) along with a more knowledgeable group (professional video-game programmers; low-SK condition) or a less knowledgeable group (elementary-school students; high-SK condition). As a manipulation check, participants then rated their subjective knowledge of video games relative to the comparison group (see Table 1).

Participants were informed that the price range of the listed video games was $15 to $20. A plot description and consumer rating of each video game was provided. The average rating and standard deviation of the ratings were held constant across the choice sets (small and large). After choosing one video game, participants were told that 2 participants would be randomly selected to receive additional compensation—either the game they had chosen or $5. Participants indicated which of these they would prefer if they were selected. It was noted that they would not be able to change their preference later. Participants then rated their absolute subjective knowledge (as opposed to subjective knowledge relative to the comparison group) of video games (two items; see Table 1; α = .86). The relative subjective knowledge measure was taken at the beginning of the study, as a manipulation check, before a choice was made; the measure of absolute subjective knowledge was measured at the end of the study, after participants had made their choice of video game. Finally, participants filled out two items as a check for choice-set size manipulation (see Table 1; α = .72).

Results and discussion

Manipulation checks

Participants’ ratings of their video-game subjective knowledge relative to the comparison group were lower in the low-SK condition (M = 2.39) than in the high-SK condition (M = 3.73), t(143) = −5.7, p < .0001. The choice set was perceived as smaller in the small-set-size condition (M = 3.39) than in the large-set-size condition (M = 4.45), t(141) = −4.4, p < .0001.

Choice

A logistic regression revealed a significant SK Condition × Choice-Set Size interaction, χ²(1, N = 145) = 9.4, p = .002: Low-SK participants were more likely to select the video game over the $5 when the choice set was large (64%) than when it was small (36%; Fig. 3), χ²(1, N = 75) = 5.9, p = .02. In contrast, high-SK participants were less likely to select the video game over the $5 when the choice set was large (29%) than when it was small (53%), χ²(1, N = 70) = 3.9, p = .047. No main effect was found for either SK condition, χ²(1, N = 145) = 0.2, p = .27, or choice-set size, χ²(1, N = 145) = 0.04, p = .81.

Fig. 3.

Results from Study 3: the percentage of participants who chose the video game rather than cash as a function of subjective-knowledge (SK) condition and choice-set size.

Rated absolute subjective knowledge

A two-way analysis of variance (ANOVA) conducted on rated absolute subjective knowledge revealed a pattern similar to that in the choice data. Specifically, there was a significant SK Condition × Choice-Set Size interaction, F(1, 141) = 18.5, p < .0001: In the low-subjective-knowledge condition, participants rated their absolute subjective knowledge higher when the choice set was large (M = 5.62) than when it was small (M = 4.61; Fig. 4), F(1, 141) = 7.7, p = .006. Conversely, in the high-subjective knowledge condition, participants rated their absolute subjective knowledge lower when the choice set was large (M = 2.93) than when it was small (M = 4.17), F(1, 141) = 10.9, p = .001. No main effect of choice-set size was found, F(1, 141) = 0.2, p = .65. Participants in the low-SK condition rated their absolute subjective knowledge higher (M = 5.11) than did participants in the high-SK condition (M = 3.55), F(1, 141) = 36.1, p < .0001.

Fig. 4.

Results from Study 3: mean rated absolute subjective knowledge (SK) as a function of SK condition and choice-set size.

We performed a moderated mediation analysis using a bootstrapping procedure (Preacher & Hayes, 2008). The analysis revealed significant indirect but opposite effects of absolute subjective knowledge on willingness to purchase: In the low-SK condition, participants who chose from a large set rather than a small set rated their absolute subjective knowledge higher and were thus more likely to prefer the video game to the cash, b = 0.31, 95% confidence interval (CI) = [0.01, 0.06]. Conversely, in the high-SK condition, participants who chose from a large set rather than a small set rated their absolute subjective knowledge lower and were thus less likely to prefer the video game, b = −0.34, 95% CI = [−0.08, −0.02].

The results of Study 3 replicated the moderating role of subjective knowledge on the relationship between choice-set size and willingness to purchase, and they provide evidence that this effect is mediated by rated absolute subjective knowledge.

General Discussion

Previous research has indicated that whereas decision makers find a greater number of choice options more appealing, the provision of choice often has adverse effects on willingness to purchase, confidence in choice, and satisfaction with the selected option. In the studies reported here, we examined the role of subjective knowledge in moderating choice overload. We used different choice sets and different subjective-knowledge manipulations, and obtained converging evidence that low-SK people are more willing to purchase when choosing from a large set of options than when choosing from a small set of options (a pattern consistent with the notion of “more is better”) and that high-SK people are less willing to purchase when choosing from a large set of options than when choosing from a small set of options (a pattern consistent with previous evidence of choice overload). The results of Study 2 further showed that this pattern is influenced by the informativeness of the descriptions of the available options and occurs regardless of decision makers’ level of actual knowledge. Study 3 revealed that postchoice subjective knowledge (rated absolute subjective knowledge) mediates the interactive effect of prechoice subjective knowledge and set size on choice. Together, our findings suggest that subjective knowledge may play an important role in determining ideal size for choice sets. They suggest that more options should be provided in domains in which people often feel ignorant (e.g., wine), whereas fewer options should be offered in domains in which people tend to feel knowledgeable (e.g., soft drinks).

At first blush, our results may seem inconsistent with findings reported by Iyengar, Huberman, and Jiang (2004). Using real choice data, these researchers found that participation rates in 401(k) retirement saving plans fall as the number of options increases. Given that employees tend to have both low actual knowledge and low subjective knowledge regarding 401(k) plans (Choi, Laibson, & Madrian, 2005; Lusardi & Mitchell, 2007), our model suggests that employees should benefit from the provision of more choice. Study 2 offers an explanation for this apparent inconsistency: Previous research has shown that even though 401(k) participants choose their plans and have some experience with the plans, they do not understand the basic features, benefits, and consequences of these plans, presumably because this information is too complex (Choi et al., 2005; Lusardi & Mitchell, 2007). As the results of Study 2 suggest, if employees’ subjective knowledge regarding 401(k) plans is low, and if the information about the options provided is complex and therefore choosing a 401(k) plan does not increase employees’ subjective knowledge, then employees should not benefit from larger choice sets. In fact, if the complexity of this product information diminishes subjective knowledge, employees may be deterred by more choice. Future research may determine whether uninformative and complicated product information have similar effects on choice overload.

Footnotes

Acknowledgements

We thank the Associate Editor and three anonymous reviewers for their constructive input. We also thank Shai Danziger, David Tannenbaum, and Daniel Oppenheimer for their valuable comments on earlier drafts.

Declaration of Conflicting Interests

The authors declared that they had no conflicts of interest with respect to their authorship or the publication of this article.

Funding

This research was supported in part by Marie Curie International Reintegration Grant PIRG06-GA-2009-252592 (to L. Hadar).

Notes

References

Alba

J. W.

Hutchinson

W. J.

(1987). Dimensions of consumer expertise. Journal of Consumer Research, 13, 411–454.

Alba

J. W.

Hutchinson

W. J.

(2000). Knowledge calibration: What consumers know and what they think they know. Journal of Consumer Research, 27, 123–156.

Arrow

K. J.

(1959). Rational choice functions and orderings. Economica, 26, 121–127.

Bearden

W. O.

Hardesty

D. M.

Rose

R. L.

(2001). Consumer self-confidence: Refinements in conceptualization and measurement. Journal of Consumer Research, 28, 121–134.

Bellenger

D. N.

Korgaonkar

P. K.

(1980). Profiling the recreational shopper. Journal of Retailing, 56, 77–92.

Botti

Iyengar

S. S.

(2004). The psychological pleasure and pain of choosing: When people prefer choosing at the cost of subsequent outcome satisfaction. Journal of Personality and Social Psychology, 87, 312–326.

Chernev

(2003a). Product assortment and individual decision processes. Journal of Personality and Social Psychology, 85, 151–162.

Chernev

(2003b). When more is less and less is more: The role of ideal point availability and assortment in consumer choice. Journal of Consumer Research, 30, 170–183.

Choi

J. J.

Laibson

Madrian

B. C.

(2005). $100 bills on the sidewalk: Suboptimal investment in 401(k) plans (NBER Working Paper No. 11554). Retrieved from the National Bureau of Economic Research Web site: http://www.nber.org/papers/w11554

10.

Diehl

Poynor

(2010). Great expectations?! Assortment size, expectations and satisfaction. Journal of Marketing Research, 47, 312–322.

11.

Fernandes

Lynch

J. G.

Jr. Netemeyer

R. G.

(2014). Financial literacy, financial education, and downstream financial behaviors. Management Science. Advance online publication. doi:10.1287/mnsc.2013.1849

12.

Fox

C. R.

Weber

(2002). Ambiguity aversion, comparative ignorance, and decision context. Organizational Behavior and Human Decision Processes, 88, 476–498.

13.

Hadar

Sood

Fox

C. R.

(2013). Subjective knowledge in consumer financial decisions. Journal of Marketing Research, 50, 303–316.

14.

Iyengar

Huberman

Jiang

(2004). How much choice is too much? Contributions to 401(k) retirement plans. In Mitchell

O. S.

Utkus

(Eds.), Pension design and structure: New lessons from behavioral finance (pp. 83–95). Oxford, England: Oxford University Press.

15.

Iyengar

Lepper

M. R.

(2000). When choice is demotivating: Can one desire too much of a good thing? Journal of Personality and Social Psychology, 79, 995–1006.

16.

Kahn

B. E.

(1995). Consumer variety-seeking among goods and services: An integrative review. Journal of Retailing and Consumer Services, 2, 139–148.

17.

Lichtenstein

Fischhoff

Phillips

L. D.

(1982). Calibration of probabilities: The state of the art to 1980. In Kahneman

Slovic

Tversky

(Eds.), Judgment under uncertainty: Heuristics and biases (pp. 306–334). Cambridge, England: Cambridge University Press.

18.

Lusardi

Mitchell

O. S.

(2007). Financial literacy and retirement preparedness: Evidence and implications for financial education. Business Economics, 42, 35–44.

19.

McAlister

(1982). A dynamic attribute satiation model of variety-seeking behavior. Journal of Consumer Research, 9, 141–150.

20.

Mogilner

Rudnick

Iyengar

S. S.

(2008). The mere categorization effect: How the presence of categories increases choosers’ perceptions of assortment variety and outcome satisfaction. Journal of Consumer Research, 35, 202–215.

21.

Moorman

Diehl

Brinberg

Kidwell

(2004). Subjective knowledge, search location, and consumer choice. Journal of Consumer Research, 31, 347–356.

22.

Park

C. W.

Gardner

M. P.

Thukral

V. K.

(1988). Self-perceived knowledge: Some effects on information processing for a choice task. American Journal of Psychology, 101, 401–424.

23.

Park

C. W.

Mothersbaugh

D. L.

Feick

(1994). Consumer knowledge assessment. Journal of Consumer Research, 21, 71–82.

24.

Parker

A. M.

Bruine

Bruin

Yoong

Willis

(2012). Inappropriate confidence and retirement planning: Four studies with a national sample. Journal of Behavioral Decision Making, 25, 382–389.

25.

Preacher

K. J.

Hayes

A. F.

(2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40, 879–891.

26.

Radecki

C. M.

Jaccard

(1995). Perceptions of knowledge, actual knowledge, and information search behavior. Journal of Experimental Social Psychology, 31, 107–138.

27.

Raju

P. S.

Lonial

S. C.

Mangold

W. G.

(1995). Differential effects of subjective knowledge, objective knowledge, and usage experience on decision making: An exploratory investigation. Journal of Consumer Psychology, 4, 153–180.

28.

Scheibehenne

Greifeneder

Todd

P. M.

(2010). Can there ever be too many options? A meta-analytic review of choice overload. Journal of Consumer Research, 37, 409–425.

29.

Schwarz

(2004). Metacognitive experiences in consumer judgment and decision making. Journal of Consumer Psychology, 14, 332–348.

30.

See

K. E.

(2009). Reactions to decisions with uncertain consequences: Reliance on perceived fairness versus predicted outcomes depends on knowledge. Journal of Personality and Social Psychology, 96, 104–118.

31.

Urbany

Dickson

P. R.

Wilkie

W. L.

(1989). Buyer uncertainty and information search. Journal of Consumer Research, 16, 208–215.

32.

Wood

S. L.

Lynch

J. G.

Jr. (2002). Prior knowledge and complacency in new product learning. Journal of Consumer Research, 29, 416–426.