Online dating choice satisfaction: The joint effect of choice overload and the tendency to maximize

(1)

Online dating choice satisfaction: The joint effect of choice

overload and the tendency to maximize

Master Thesis

T.R. Langhorst BSc. – 2151340 Economic & Consumer Psychology

Leiden University

Supervisor: Dr. H.A.H. Ruigendijk December 7th, 2018

Abstract

The purpose of this research was to investigate the influence of choice overload and the tendency to look for the best available option (maximizing) on choice satisfaction in online dating. In an online experiment, simulating an online dating website, a decrease in satisfaction when presented with choice overload in combination with a tendency to maximize was expected. Both were not found to be influential. However, larger choice sets might provide different outcomes. With more people using online dating, the effect of the number of possible dating partners and the influence of personal differences should be investigated carefully. This research provides insight into the conditions in which the consequences of choice overload are observable, and suggests possibilities for future research.

(2)

In recent years, the quest for love and meeting new people has partly transferred from the real world to the World Wide Web in the form of online dating. Of all Americans, 11 per cent indicates they have used online dating websites or applications and 29 per cent of Americans know someone who has used online dating to find a romantic partner (Masden & Edwards, 2015). Websites like Match.com or eHarmony.com have shown to be successful in helping people find a romantic partner (Finkel, Eastwick, Karney, Reis & Sprecher, 2012). Such websites provide search tools, which help members to find online profiles of other users and enable them to contact these users (Yang & Chiou, 2010). Every dating platform provides its own speciality. For example, PlentyOfFish claims their website will give you access to 145 million monthly visitors, and 2,5 million of them have conversations every day. The company even advertises the fact there are no other dating websites with more singles looking for new people to meet. Although this might sound like a good thing (a big sea full of fish to choose from), having millions of options might also create some difficulties. Having so many options that you can not choose could eventually lead to being less satisfied with your choice. The effect in which having too many options decreases choice satisfaction is also known as the choice overload effect (Scheibehenne, Greifeneder & Todd, 2009).

The goal of the current study is to investigate whether having a lot of options in an online dating setting is equally detrimental for everyone, or that individual differences have an influence on the effect. More specifically, this research will focus on the extent to which people tend to maximize their outcomes. The literature

distinguishes the tendency to either search for the best available option (to maximize), or settle for an acceptable option in which a good option is good enough (to satisfice) (Scheibehenne, Greifeneder & Todd, 2009). Based on the literature, we expect

(3)

mazimizers to suffer more from the choice overload effect in comparison to satisficers (Schwartz, Ward, Monterosso, Lyubomirsky, White, & Lehman, 2002; Dar-Nimrod, Rawn, Lehman, D, & Schwartz, 2009; Scheibehenne, Greifeneder & Todd, 2009).

The Choice Overload Effect

The choice overload effect has been used widely to explain people’s purchases in a

variety of choice situations such as chocolates and jam (Iyengar & Lepper, 2000). Iyengar and Lepper (2002) found that people like having lots of options, but that this also increases choice difficulties. In their study, they found people were more

interested in a stand with 24 types of jam compared to a stand with only 6 types of jam. However, the participants bought more from the stand with only 6 types of jam. Furthermore, Schwartz et al. (2002) concluded there is a negative correlation between the amount of options to choose from and how satisfied people are with their choices. This effect has been found in similar studies (Park & Jang, 2013; Mogilner, Rudnick, & Iyengar, 2008), which offered people either a choice overload or just a few options.

The effect of choice overload has also been found in the context of online dating. D’Angelo and Toma (2017) found lower satisfaction levels when participants were presented with 24 dating options compared to participants who were presented with only six dating. Moreover, after one week of thinking, the participants who were presented with 24 dating options were more likely to take the opportunity to change their dating choice made one week earlier compared to participants who were presented with six dating options.

A variety of explanations have been offered for why the choice overload effect occurs. For example, when options increase in a certain category, the differences between options decrease. At the same time, the amount of information increases

(4)

(Fasolo, McClelland & Todd, 2007). Moreover, having a large amount of options makes an exhaustive comparison in which every option is considered less desirable, taking into account the time and effort that is needed to make the best choice. However, not taking every option into account could lead to the feeling that the optimal option is missed out on (Schwartz et al, 2002).

Another cause for the choice overload effect involves the increase of

expectations when the amount of options increases. In a large choice set with similar options, the chosen option might not live up to the expectations. This could ultimately decrease satisfaction (Diehl & Poynor, 2010).

Also, when options increase, people might engage more in counterfactual thinking. Counterfactual thinking is a process in which people start thinking about what could have been instead of thinking about what they have (Hafner, White & Handley, 2012). In large choice sets, engaging in counterfactual thinking can be disadvantageous, due to the large range of other possibilities. This negative evaluation of the outcome could therefore decrease satisfaction (Hafner et al., 2012).

And last, a possible explanation for why people are less satisfied when options increase is anticipated regret; a feeling that is caused by the knowledge that one must live with the outcomes of one’s choices. In large choice sets, the outcome can very well be lower compared to the best possible outcome, and could therefore cause anticipating regret (Schwartz et al., 2002). Regret has found to be correlated to satisfaction (Tsiros & Mittal, 2000). Schwartz et al. (2002) found that when options increase, so does anticipated regret, which could consequently lead to a decrease in satisfaction.

(5)

Maximization tendency in relation to Choice overload

Research shows when amount of options increase, satisfaction decreases. The literature however points out that maximizers are also less satisfied when offered large choice sets. The mechanisms underlying these effects are similar to each other. Therefore, maximization tendency could be a moderator to the choice overload effect. For example, maximizers prefer large choice sets, but also have problems choosing between large amounts of options (Schwartz et al., 2002). Schwartz et al. (2002) suggested that maximizers tend to form higher expectations compared to satisficers. When expectations are higher, they are also more difficult to be fulfilled, and when expectations are not met, satisfaction might ultimately decrease (Diehl & Poynor, 2010). Choice overload, as mentioned earlier, also creates higher expectations (Diehl & Poynor, 2010). Therefore, because maximizers tend to form higher expectations in general, it is likely that maximizers are more vulnerable for the choice overload effect compared to satisficers.

Not only expectations influence maximizers and satisficers differently, counterfactual thinking also does. Having a maximizing decision-making orientation is related to engaging in more upward counterfactual thinking, which is comparing your own outcome to a better or even the best possible outcome (Leach & Patall, 2013). Compared to the best possible outcome, the actual outcome is almost always a disappointment. Therefore, the upward counterfactual thinking will ultimately

decrease satisfaction (Leach & Patall, 2013). In choice overload, counterfactual thinking also occurs. Because maximizers tend to engage in more counterfactual thinking compared to satisficers, this could indicate that maximizers are more vulnerable to the choice overload effect.

(6)

Maximizers do not only engage in more counterfactual thinking, but also anticipate more on how much they will regret their decisions (Schwartz et al, 2002). Because alternatives may overwhelm maximizers, they anticipate regretting

opportunities they missed out on, and are unhappy about unmet expectations. This leads maximizers to be afraid of making choices and be less satisfied afterwards (Chowdhury, Ratneshwar & Mohanty, 2009). Anticipated regret also plays an active role in the choice overload effect. Once again, because maximizers tend to anticipate regret in large choice sets, this could indicate they are more sensitive to the choice overload effect compared to satisficers.

There is also more direct evidence that having too many options is

disadvantageous for maximizers, but in which the tested variable was not satisfaction. Yang and Chiou (2010) tested the influence of choice overload in an online dating setting, and controlled for the influence of a high maximization tendendy. However, instead of testing choice satisfaction, they looked at the quality of the decisions. Because maximizers tend to always look for the best available option, they put more time and effort into looking for alternatives compared to satisficers. Therefore, when there is no choice overload, maximizers do get the best objective outcomes (Polman, 2010). However, when options increase and more information should be processed in order to get the best result, maximizing outcomes becomes more difficult. Yang and Chiou (2010) concluded that maximizers make poorer decisions due to reduced selectivity when offered a choice overload. Wu and Chiou (2009) argued that due to excessive searching, reduced cognitive resources were the cause of the worse decisions.

(7)

The present study

In the present study, we investigated the effects of choice overload in online dating on satisfaction and the moderating effect of the tendency to maximize or satisfice. The participants were presented with either six online profiles (no choice overload condition), or 24 online profiles (choice overload condition). The participants were asked to choose one of the persons, which they would like to have a date with. Afterwards choice satisfaction was measured. After one week, the participants were asked to revisit the website and answer how satisfied they were with the choice they made one week earlier. Also the tendency to maximize or satisfice was measured after second measurement. Given the mentioned literature about choice overload and the tendency to maximize or satisfice, the following hypotheses were developed.

H1: Choice overload in an online dating context will lead to lower choice satisfaction compared to no choice overload.

H2: The tendency to maximize/satisfice moderates the choice overload effect. That is, the joint effect of choice overload with high tendencies of maximization will lead to a decrease in satisfaction compared to no choice overload and low levels of

maximization tendency.

Both hypotheses will be tested on the measurement of satisfaction one week after the choice was made and, because we expect the effects to develop over time (D’Angelo & Toma, 2017). The hypotheses will also be tested on the difference score between the first measurement of satisfaction and the measurement one week later.

(8)

Method

Participants & Recruitment

The participants (N = 94, age M = 34.19, SD = 2.99) were women between the age of 30 and 40, single and heterosexual. Participants were recruited using Prolific, an online platform for participant recruitment. The participants in the choice overload condition received $3,25 for their participation (30 minutes). They received $1,25 after the first part of the study and $2 for their participation in the follow up measurement one week later. Participants in the no choice overload condition

received $2,17 for their participation (20 minutes). They received $0,83 after the first part of the study and $1,34 for their participation in the follow up measurement one week later.

Procedure and Design

The current study had a two-group between (Choice set size; six or 24 stimuli) and within (time 1 and time 2) experimental design with the tendency to maximize or satisfice as a moderator and choice satisfaction as the dependent variable. The

participants were randomly assigned to either the large choice set condition (choice overload) or the small choice set condition (no choice overload) using randomization software. The choice set sizes were established after reviewing other choice overload studies. In the meta-analysis by Chernev, Böckenholt and Goodman (2015), multiple choice overload studies are presented, in which most of them use a small choice set of six different items, and a large choice set of 24 different items. D’Angelo and Toma (2017) also used six and 24 options as their small and large choice sets in their study about choice overload in online dating. Therefore, the current study will also present

(9)

the participants with six options in the no choice overload condition, and 24 dating options in choice overload condition.

Participants took part in two online sessions, with one week between these sessions. D’Angelo and Toma (2017) argued that the moment in which choice

overload effect occurs is later for choices that are ‘high-stake’. Online dating, in their opinion, fits these characteristics. Accordingly, the choice overload effect and with it a decrease in satisfaction will only be visible after a period of one week, when clearer attitudes have been formed. For that reason, in the present study there will also be a follow-up measurement of satisfaction at the second session.

During the first session, the participants signed the informed consent, in which they agreed to the terms of the current study (Appendix 2). Next, participants read a cover story in which they were told they were testing a new dating website in which they will be looking for a possible dating partner. Next, participants answered a short list of demographic questions and were asked to fill in a questionnaire about attitudes towards online dating. Subsequently, they were presented with their possible matches (either 6 or 24 profiles). Participants were asked to examine each possible match before they selected their potential dating partner. After their selection, participants were asked to answer seven questions about their choice satisfaction, similar to the study by D’Angelo and Toma (2017). During the second session (one week later), the

participants were again presented with every profile they had seen one week earlier, and were asked to answer the same questions about their choice satisfaction.

Subsequently, they were asked to fill in a 13-item maximization scale (Schwartz et al, 2002) about their tendency to maximize or satisfice and some additional covariate questions. Additionally, participants were asked to recall whether they had seen either 6 or 24 profiles one week earlier. This question was used as a manipulation check,

(10)

and participants who answered this question incorrectly were excluded from the analyses. Besides these questions, they were also asked to answer six questions that functioned as covariates. After these questions, the participants were debriefed.

Online profiles

The online dating system was developed specifically for this experiment. The set-up was comparable to the dating app ‘Happn’. The presented profiles contained up to three photos, a small list of basic info such as age and interests, and a small textbox with personal information. The profiles contained pictures from volunteers who consented to the terms of this study (Appendix 3). All volunteers were male, between the age of 30 and 40. The pictures that were used for their profiles were obtained from their Facebook profile.

Because an obvious best choice in a choice set can cause the choice overload effect to not emerge (Hsee & Leclerc, 1998; Chernev, 2003), this study controlled for obvious best choices by asking participants whether their choice was the obvious best one. When a person’s profile was chosen as the obvious best choice more than five times, this profile was excluded from the study. However, none of the profiles was chosen as the obvious best choice more than five times.

Additionally, because the literature pointed out attractiveness as a main choice determinator (Finkel & Eastwick, 2008), it was important that in both conditions (6 or 24 profiles), the attractiveness was equal, using the same evaluation model as

D’Angelo & Tomas (2017). The profiles were evaluated on attractiveness by

unacquainted observers in a pre-test. Every profile was rated on attractiveness on a two-item questionnaire (i.e. “How attractive is this online dater?” , “How willing would you be to date this online dater?”) on a scale from 1 to 10. After the pre-test,

(11)

the groups with either 24 or six profiles were formed. The small choice set, with six profiles (M = 4.64, SD = 2.46), was formed with profiles that were on average as attractive as the large choice set (M = 4,74, SD = 2,39). The six profiles of the small choice set were also used in the large choice set (24 profiles). In other words, the large choice set also contained the profiles of the small choice set.

Materials

Participants were given a cover story in which they were told this study was part of testing a spin off from a popular dating website in which they will be looking for a possible dating partner. The full cover story can be found in Appendix 1.

For the dependent variable a 7-item choice satisfaction inventory was used that was similar to the research by D’Angelo & Toma (2017). This scale contained

questions like “How much do you like the individual whose profile you selected?”, “How satisfied are you with the profile you chose?”, and “How much are you looking forward to contacting this individual?”. Each question was answered by on a 7-point

Likert scale ranging from 1 (not at all) to 7 (extremely).

For the measurement of tendency to maximize or satisfice the 13-item

maximization scale by Schwartz (2004) was used. This scale contained questions like “No matter how satisfied I am with my job, it's only right for me to be on the lookout for better opportunities”, and “I treat relationships like clothing: I expect to try a lot on before finding the perfect fit”. Again, each question was answered by on a 7-point

Likert scale ranging from 1 (completely disagree) to 7 (completely agree). A high score on these questions indicated a higher tendency to maximize, whereas a low score indicated a tendency to satisfice.

(12)

Additionally, we asked participants on how many profiles they clicked during the study. Participants were not obligated to look at every profile, so asking on how many profiles participants clicked could provide interesting insights.

Covariates

A list of covariates was obtained using the same questions as D’Angelo & Toma

(2017). These were measured to control for any influence they may have on choice satisfaction.

a. Tendency for romantic idealization (e.g. ‘Do you believe in soulmates?’). Romantic idealization is a reason for more positive illusions, which could lead to

stronger satisfaction (Murray, Holmes, & Griffin, 1996)  

b. Online dating experience (e.g. ‘Have you ever dated someone you met through online dating?’). Online daters with experience in online dating might be more

comfortable with the tools (Sautter, Tippett, Morgan, 2010).

c. Attitudes towards online dating (‘Would you be willing to use online dating in the future?’). A very negative stigma towards online dating could influence the

outcomes (Cali, Coleman, & Campbell, 2013). 

d. Online dating efficacy (‘I can use online dating to get what I want.’) When people are not comfortable using online dating, they might not be responsive to manipulation. e. Previous relationship experience (‘How many committed romantic relationships

have you had?’). Lots of relationship experience can influence the way new

partners are observed (Furman, Brown, & Feiring, 1999).

f. Potential partner image (‘How clear can you picture what your future partner should look like?’). When people can picture their future partner very clearly,

(13)

Results

To be sure the choice overload manipulation caused people to look at more profiles, a one-way ANOVA was run, with the amount of profiles participants clicked on as dependent variable, and the condition (choice overload or no choice overload) as independent variable. The F-test was significant, F(1,93) = 6.230, p = .014, ηp2 = .063,

with the participants in the choice overload condition (M = 6.35, SD = 6.28) clicking on more profiles compared to the no choice overload condition (M = 4.16, SD = 1.88). This was in line with our expectations.

The hypotheses predicted outcomes on satisfaction levels at the second measurement and the difference score between the first and second measurement. However, to ensure no effects emerged at the first measurement, a linear regression analyses was run, with choice satisfaction at the first measurement as dependent variable. The two independent variables and the interaction term were used as between-subject factors together with all covariates. The regression was

non-significant, F(10, 93) = 1.238, p = .279. No main effect was found for choice overload (ß = -1.44, p = .481), with no significant difference between participants with six options (M = 5.11, SD = .93) and participants with 24 options (M = 5.05, SD = .82). No main effect was found for maximization tendency (ß = -.055, p = .634) and no moderating effect was found for maximization tendency in combination with choice overload (ß = -.277, p = .184). Of the covariates, none were significant. For this reason, no support was found for an effect of choice overload at the first

measurement, which is in line with the findings by D’Angelo and Toma (2017).

The first hypothesis predicted a significantly lower satisfaction level for the choice overload condition compared to the no choice overload condition at the second measurement and for the satisfaction difference score. The difference score was

(14)

computed by subtracting the satisfaction score of the first measurement from the satisfaction score of the second measurement. With two linear regression analyses, in which either choice satisfaction at the second measurement (Linear regression 1) or the difference score of choice satisfaction (Linear regression 2) were the dependent variable, this hypothesis was tested. In both regressions, the same between-subject factors and covariates were used as mentioned earlier.

Linear regression 1 was non-significant, F(10,93) = 0.790, p = .638. None of the covariates were significant, all p’s > .05. The average satisfaction levels at the second measurement of the choice overload condition (M = 4.99, SD = 1..33) and the no choice overload condition (M = 5.16, SD = 1.21) did not differ significantly (ß = -.256, p = .391). Linear regression 2, with the difference score of satisfaction between the second and the first measurement as the dependent variable, also was

non-significant, F(10,93) = 0.562, p = .840. None of the covariates were non-significant, all p’s > .05. Similar to the first regression, the difference score of satisfaction of the choice overload condition (M = -0.07, SD = 1.11) and the no choice overload condition (M = 0.04, SD = 0.92) did not differ significantly (ß = -.112, p = .640). Consequently, these results do not match the predictions made in the first hypothesis.

The second hypothesis predicts lower levels of satisfaction for higher levels of maximization tendency in combination with the choice overload condition for the second measurement of satisfaction and also the difference score of satisfaction. The first linear regression showed no significant moderating effect of maximization (ß = -.192, p = .528). Also with the difference score of satisfaction, the second linear regression did not report a significant moderating effect of maximization (ß = .085, p = .726). These outcomes do not support the predictions made in the second

(15)

Exploratory Data

An extra analysis was run, with the amount of profiles clicked on as the dependent variable, the amount of profiles seen as the independent variable and the tendency to maximize as the moderator. Results indicated that the amount of profiles presented (b = 2.33, SEb = .88, β = -.506, p = .009), but not maximization tendency (b = -.497, SEb = .048, β = -.540, p < .001), is related to amount of profiles clicked on. The

interaction between amount of profiles presented and maximization tendency was however moderately significant (b = 1.181, SEb = .92, β = -.344, p = .053), suggesting that the effect of amount of profiles presented on amount of profiles clicked on depended on the level of maximization tendency. Simple slopes for the relation between choice overload and amount of profiles clicked on were tested for low (-1 SD below the mean), moderate (mean), and high (+1 SD above the mean) levels of maximization tendency. The simple slope tests revealed a significant

positive relation between amount of profiles presented and amount of profiles clicked on. However, the amount of options presented was more strongly related to amount of profiles clicked on for high levels of maximization tendency (b = 4.14, SEb = 1.41, β = .88, p = .002, 95% CI [1.54, 6.75]) than for moderate (b = 2.34, SEb = .88, β = -.54, p = .009, 95% CI [0.60, 4.08]) or lower (b = .53, SEb = 1.23, β = -.20, p = .670, 95% CI [-1.92, 2.97]) levels of maximization tendency. This means participants in the choice overload condition, in combination with higher levels of maximization

tendency, clicked on more profiles compared to lower levels of maximization tendency in the choice overload condition. This effect was not visible when choice satisfaction was used as the dependent variable.

Discussion

(16)

and its relation to satisfaction, are becoming more important as online dating is becoming more popular (Masden & Edwards, 2015). The amount of profiles

presented on a dating site differs among online dating companies, and might therefor influence the chances of success of creating a romantic relationship. But not only the amount of profiles presented, also individual differences have shown to be of

influence when making choices, with the tendency to maximize ones outcomes as a determinant of satisfaction (Yang & Chiou, 2010). This research focused on these individual differences and choice overload, investigating the influence of having too many options (choice overload) in online dating and the joint effect of the tendency to maximize and choice overload on choice satisfaction. More precisely, we expected choice overload and choice overload in combination with high tendency to maximize to have a negative effect on choice satisfaction. The results do not support these expectations. Participants who were presented with 24 dating profiles were not less satisfied with their dating choice compared to participants who were presented with only six dating profiles. Also, participants with high tendency to maximize who were presented with 24 dating profiles were not less satisfied with their dating choice compared to participants with lower tendency to maximize and who were only presented with six profiles. In other words, our results do not support a moderating effect of maximization and choice overload on choice satisfaction. Additionally, the expectation that time was an important ingredient for the emergence of choice

overload was not met. There was no main effect of choice overload and maximization tendency on the difference scores between the second and first measurement of satisfaction. Also, no interaction effect was observed.

Given the outcomes of this investigation, the current study provides interesting insights into the exact workings of the choice overload effect, and the impact of

(17)

personal differences. Starting with the choice overload effect, the results could not provide evidence for the influence of amount of options on choice satisfaction. This finding suggests the choice overload effect to be less robust than expected. A possible explanation for these findings could be that high-stake choices need bigger choice sets in the choice overload condition in order to find a difference in satisfaction. This study used a questionnaire in which participants of each condition had the possibility to look at every profile they were presented with. However, participants in the choice overload condition looked at 6.35 profiles and to 4.16 profiles in the no choice overload condition. Although the conditions were different, it is possible that looking at only 6.35 profiles on average was not enough to create a real overload. Also, the profiles created in this study were on average evaluated on attractiveness with a 4.74 on a scale of 1 to 10. To ensure overload, and more difficulties choosing the best possible option, a higher average of attractiveness could be a solution. Future research could therefore obligate participants to look at every profile and try to create a more attractive stimuli sample to ensure a choice overload.

Likewise, adding the personal differences of the satisficing/maximizing construct in combination with the choice overload into the analyses unexpectedly did not influence choice satisfaction. Earlier research suggested the theory that the effects of having a maximizing tendency were rather similar to the effects of choice overload and could therefore moderate the negative effect on choice satisfaction. The

unexpected outcome suggests the possibility that these two constructs, although still similar in the way they reduce cognitive resources to find the best possible outcome, do not operate as predicted with the given choice set sizes. It may simply be that having 24 online dating options does not cause enough depletion in cognitive

(18)

choice satisfaction. This theory is supported by Ma (2016), who found lower ‘search costs’ when searching online for the best option compared to searching offline. Larger

choice sets could therefore be needed in an online setting to find an effect.

Additionally, our exploratory data suggest that higher tendencies to maximize did not cause participants to click on more profiles with the given choice sets. However, when looked at the amount of profiles clicked on, the same exploratory analysis did find an interaction effect between the maximization tendency and choice overload. The data showed participants with higher levels of maximization tendency in

combination with choice overload to click on more profiles compared to lower levels of maximization and no choice overload. Interestingly, the study by Yang and Chiou (2010), in which they provided participants with either 40 or 80 dating options and also controlled for the influence of maximization tendency, did find an effect for choice overload and maximization. The outcomes of the exploratory data and the study by Yang and Chiou (2010) suggest the possibility that the choice set sizes used in this study did not exceed the threshold to create overload. Therefore, presenting too few options to create overload could be a reason this study did not find an interaction effect. Future research should look into the effect of choice overload in combination with maximization tendency on satisfaction with larger choice sets to examine this explanation.

In contrary to the above, D’Angelo and Toma (2017) did find an effect using the same choice sets as used in the current study. Chernev, Böckenholt and Goodman (2015), suggested experience in a particular field to influence the amount of options needed to create a choice overload effect. They suggested the theory that when people have experience, larger choice sets than normally in a particular field are needed to create a choice overload effect. As a consequence, there is the possibility the

(19)

experience in dating among participants might be lower in the study done by D’Angelo and Toma (2017) compared this study, for example because in the study done by D’Anglo and Toma (2017), college students were used as participants, whereas in this study, adults between the age of 30 and 40 were recruited. According to Smith and Alloy (2009), adults between the age of 25 and 45 are the most ‘typical’ online daters and could therefore have more experience in online dating compared to college students, which caused the adults to not experience any choice overload in both conditions. Future research should look into the impact of experience in high-stake choices to create better understanding.

Practical Implications

Dating websites nowadays use different kinds of choice set sizes to present to their customers, ranging from only four (eHarmony) to over a hundred (Plenty Of Fish). When it comes to choice overload, the current research suggests the choice overload effect is less robust than expected. However, interesting explanations have been provided as for why no effect was found. The amount of options needed to create choice overload in dating is therefore an interesting topic for future research. Larger choice sets, comparable with the study by Yang and Chiou (2010), in combination with a measurement of tendency to maximize, might provide different outcomes and give more information about the boundaries of the choice overload effect.

Limitations & Future research

This online study could not establish the quality and seriousness of the participants. It is possible the participants did not put enough effort in selecting the best possible partner, which could have influenced the results. Also, they could have

(20)

lied about their interest in online dating or their gender to be allowed to take part in this online research in order to get paid for their participation. Moreover, with only 37 participants in the choice overload condition, this research fails to create sufficient power to find an effect. Therefore, we invite future research to replicate these findings with surveillance over the participants, and sufficient participants in both conditions.

Conclusion

With a growing number of people dating online, the possibilities of meeting people through the internet become larger every year. Accordingly, this research has shown greater insight in the emergence of the choice overload effect and the influence of personal differences on the choice overload effect. This research shows future research should, to observe a choice overload effect, employ larger choice sets. By testing the choice overload effect and the joint effect of maximization and choice overload directly in an online dating setting, presenting either six or 24 options did not affect choice satisfaction. Different explanations for these findings are presented, and several interesting possibilities for future research are proposed. At this point however, the perfect amount of options and the influence of maximization tendency for ultimate satisfaction remain unknown to a world of people dating online.

(21)

References

Cacioppo, J. T., Cacioppo, S., Gonzaga, G. C., Ogburn, E. L., & VanderWeele, T. J. (2013). Marital satisfaction and break-ups differ across on-line and off-line meeting venues. Proceedings of the National Academy of Sciences, 110(25), 10135-10140.

Cali, B. E., Coleman, J. M., & Campbell, C. (2013). Stranger danger? Women’s self- protection intent and the continuing stigma of online dating. Cyberpsychology, Behavior, and Social

Networking, 16(12), 853–857.

Chernev, A. (2003). When more is less and less is more: The role of ideal point availability and assortment in consumer choice. Journal of consumer Research, 30(2), 170-183.

Chernev, A., Böckenholt, U., & Goodman, J. (2015). Choice overload: A conceptual review and meta-analysis. Journal of Consumer Psychology, 25(2), 333-358.

Chowdhury, T. G., Ratneshwar, S., & Mohanty, P. (2009). The time-harried shopper: Exploring the differences between maximizers and satisficers. Marketing Letters, 20(2), 155-167. D’Angelo, J. D., & Toma, C. L. (2017). There are plenty of fish in the sea: The effects of choice

overload and reversibility on online daters’ satisfaction with selected partners. Media Psychology, 20(1), 1-27.

Dar-Nimrod, I., Rawn, C. D., Lehman, D. R., & Schwartz, B. (2009). The maximization paradox: The costs of seeking alternatives. Personality and Individual Differences, 46(5-6), 631-635. Demir, M. (2010). Close relationships and happiness among emerging adults. Journal of Happiness

Studies, 11(3), 293-313.

Diehl, K., & Poynor, C. (2010). Great expectations?! Assortment size, expectations, and

satisfaction. Journal of Marketing Research, 47(2), 312-322. Fasolo, B., McClelland, G. H., & Todd, P. M. (2007). Escaping the tyranny of choice: When

fewer attributes make choice easier. Marketing Theory, 7(1), 13-26.

Finkel, E. J., & Eastwick, P. W. (2008). Speed-dating. Current Directions in Psychological Science, 17(3), 193-197.

Finkel, E. J., Eastwick, P. W., Karney, B. R., Reis, H. T., & Sprecher, S. (2012). Online dating: A critical analysis from the perspective of psychological science. Psychological Science in the Public Interest, 13(1), 3-66.

(22)

adolescence. Cambridge, UK: Cambridge University Press.

Hafner, R. J., White, M. P., & Handley, S. J. (2012). Spoilt for choice: The role of

counterfactual thinking in the excess choice and reversibility paradoxes. Journal of Experimental Social Psychology, 48(1), 28-36.

Hsee, C. K., & Leclerc, F. (1998). Will products look more attractive when presented separately or together?. Journal of Consumer Research, 25(2), 175-186.

Iyengar, S. S., & Lepper, M. R. (2000). When choice is demotivating: Can one desire too much of a good thing?. Journal of personality and social psychology, 79(6), 995. Leach, J. K., & Patall, E. A. (2013). Maximizing and counterfactual thinking in academic major

decision making. Journal of Career Assessment, 21(3), 414-429.

Ma, J. (2016). Does Greater Online Assortment Pay? An Empirical Study Using Matched Online and Catalog Shoppers. Journal of Retailing, 92(3), 373-382.

Masden, C., & Edwards, W. K. (2015). Understanding the role of community in online dating. In Proceedings of the 33rd annual ACM conference on human factors in computing systems(pp. 535-544). ACM.

Mogilner, C., Rudnick, T., & 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(2), 202-215.

Moser, C., Phelan, C., Resnick, P., Schoenebeck, S. Y., & Reinecke, K. (2017, May). No

Such Thing as Too Much Chocolate: Evidence Against Choice Overload in E-Commerce. In Proceedings of the 2017 CHI Conference on Human Factors in Computing Systems (pp. 4358-4369). ACM.

Murray, S. L., Holmes, J. G., & Griffin, D. W. (1996). The self-fulfilling nature of positive illusions in romantic relationships: love is not blind, but prescient. Journal of Personality and Social

Psychology, 71, 1155–1180.

Park, J. Y., & Jang, S. S. (2013). Confused by too many choices? Choice overload in tourism. Tourism Management, 35, 1-12.

Polman, E. (2010). Why are maximizers less happy than satisficers? Because they maximize positive . and negative outcomes. Journal of Behavioral Decision Making, 23(2), 179-190.

(23)

United States. Social Science Quarterly, 91, 554–575.

Scheibehenne, B., Greifeneder, R., & Todd, P. M. (2009). Can there ever be too many options? A meta-analytic review of choice overload. Journal of Consumer Research, 37(3), 409-425.

Schwartz, B. (2000). Self-determination: The tyranny of freedom. American psychologist, 55(1), 79.

Schwartz, B. (2004). The tyranny of choice. Scientific American, 290(4), 70-75.

Schwartz, B., Ward, A., Monterosso, J., Lyubomirsky, S., White, K., & Lehman, D. R. (2002). Maximizing versus satisficing: Happiness is a matter of choice. Journal of personality and social psychology, 83(5), 1178.

Smith, J. M., & Alloy, L. B. (2009). A roadmap to rumination: A review of the definition, assessment, and conceptualization of this multifaceted construct. Clin- ical Psychology Review, 29, 116–

128.

Tsiros, M., & Mittal, V. (2000). Regret: A model of its antecedents and consequences in consumer decision making. Journal of Consumer Research, 26(4), 401-417.

Wu, P. L., & Chiou, W. B. (2009). More options lead to more searching and worse choices in finding partners for romantic relationships online: An experimental study. CyberPsychology & Behavior, 12(3), 315-318.

Yang, M. L., & Chiou, W. B. (2010). Looking online for the best romantic partner reduces decision quality: The moderating role of choice-making strategies. Cyberpsychology, Behavior, and Social Networking, 13(2), 207-210.

(24)

Appendix 1 - Cover story

We are developing a specific spin off dating program designed specifically for large numbers of people. In this dating program, you will be looking for a possible dating partner. For our pilot program we recruited single males between the age of 30 and 40 and are now recruiting females of the same age group for our pilot study. The participants will have to evaluate male profiles, and are told that at the end they can make a profile of their own to make the experiment trust worthier. Participants are not told that if you get a match with one of the single men you will be given contact details to get in touch with each other. The participants will be compensated for their additional feedback that will be used to optimize the platform before it will be launched in association with some large organizations.

(25)

Appendix 2 - Informed consent participants

Please read the following information carefully. This Prolific session consists of two parts:

1. Testing a new online dating website. For this test we need your participation now (20 minutes/10 minutes), and again in a week (5 minutes). Because we are aiming for a realistic evaluation of the website, if you have no intention to meet a romantic partner please do not participate in the current Prolific session.

2. A study on decision behavior. For this study you will fill in a questionnaire on decision related behaviors, immediately after testing the dating website today (5 minutes).

For the test of the new dating website, you will first be asked to answer some

questions about demographics (e.g. age, gender, etc.) and about your attitudes towards and experience with online dating. After that, you will start testing the dating website. You will screen 24 (6) profiles containing pictures and a description of members of the dating website. After you have screened the profiles you will be asked to select a potential dating partner. We will then ask you to evaluate the website and the

potential dating partner you have selected. A week later, we will ask you some short questions to evaluate the website once more and you will be able to contact your selected potential dating partner in case of a match.

You will be paid for your participation after you have completed the second test of the dating website (in a week from now). When you do not complete the second test of the dating website you will not be paid.

If you agree to participate, please be aware that you are free to withdraw at any point throughout the duration of the experiment.

All information you provide is anonymized and will not be associated with your name. The anonymized information you provide will be combined with information from other participants and then used for scientific analysis. All information will be kept confidential and your name will not be associated with any research findings. If for any reason during this study you do not feel comfortable, you may exit and your information will be discarded.

The test of the dating website and the study on decision behavior are both projects from Leiden University. If you have questions regarding the project, please feel free to contact us through email: Hester Ruigendijk, h.a.h.ruigendijk@fsw.leidenuniv.nl Statement:

- I have read the provided information for participants. I could ask additional

questions. My questions have been answered adequately. I have had sufficient time to decide whether or not I participate.

- I am aware that participation is completely voluntary. I know that I can decide at any moment not to participate or to stop. I do not need to provide a reason for that. - My responses are processed anonymized.

- I give consent to use my data for the purposes that are mentioned in the provided information

Name: Date:

I consent to participating in this study. 0Yes

(26)

Appendix 3 - Informed consent stimuli volunteers

Wij zijn op zoek naar mannen tussen de 30 en 40 jaar die bereid zijn enkele van hun facebook foto’s beschikbaar te stellen als onderdeel van ons afstudeeronderzoek. Wij doen onderzoek naar online dating en de twee tot vier foto’s zullen gebruikt worden om een fictief datingprofiel te maken. In dit datingprofiel wordt fictieve informatie, zoals een verzonnen naam, hobby’s en baan aan de foto toegevoegd. Dit betekent dus dat er geen echte naam of andere informatie over jou aan je foto gekoppeld zal worden.

Voorafgaand aan het onderzoek zullen de foto’s eerst op een schaal van 1-10 op aantrekkelijkheid worden beoordeeld in een pre-test door vrouwen tussen de 30-40 jaar. De beoordelingen op aantrekkelijkheid zullen alleen worden gebruikt zodat wij in staat zijn de foto’s te beoordelen op gemiddelde aantrekkelijkheid en zullen aan niemand anders dan de onderzoekers bekend zijn of worden gemaakt.

Vervolgens zullen de deelnemers aan het datingonderzoek een potentiële datingpartner kiezen uit de verschillende fictieve datingprofielen. Dit

matchingsproces is geheel fictief en puur voor de doeleinden van het onderzoek. Je zal dus nooit daadwerkelijk in contact worden gebracht met een vrouw die jouw fictieve datingprofiel selecteert. De foto zal alleen zichtbaar zijn voor deelnemers aan het onderzoek (vrouwen tussen de 30 en 40 jaar) en voor niemand anders.

• Ik heb de informatiebrief voor het beschikbaar stellen van mijn foto gelezen. Ik had voldoende tijd om te beslissen of ik meedoe.

• Ik weet dat meedoen helemaal vrijwillig is. Ik weet dat ik zonder opgave van redenen kan weigeren mijn foto’s beschikbaar te stellen.

• De foto’s zullen niet aan mijn persoonlijke informatie gekoppeld worden. • Ik geef toestemming om mijn foto’s te gebruiken, voor de doelen die in

de informatiebrief staan.

Naam: Datum: E-mailadres:

Ik geef toestemming om mijn persoonlijke foto’s zoals die op facebook beschikbaar zijn te gebruiken voor dit onderzoek.

0Yes 0No

Het onderzoek wordt uitgevoerd onder coördinatie van dr. Hester Ruigendijk, Universiteit Leiden (h.a.h.ruigendijk@fsw.leidenuniv.nl). Voor vragen, klachten of andere informatie kun je bij haar terecht.

(27)

Appendix 4; Informed consent pre-test

Please read the following information carefully. For the current study, we would like to ask you to assess the attractiveness of 24 faces. The current research is exclusively focused on your opinion. Therefore, there are no right or wrong answers.

You will be paid for your participation after you have assessed the 24 faces.