• No results found

“the old give and take… and take”

N/A
N/A
Protected

Academic year: 2021

Share "“the old give and take… and take” "

Copied!
45
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

REVISITING THE PRINCIPLE OF RECIPROCITY:

DOES STRESS MAKE US MORE SUSCEPTIBLE TO COMPLY WITH A REQUEST AFTER BEING OBLIGED WITH A CONCESSION?

“the old give and take… and take”

— Robert Cialdini

by

Cas de Carpentier Wolf June 15th, 2017

(2)

REVISITING THE PRINCIPLE OF RECIPROCITY:

DOES STRESS MAKE US MORE SUSCEPTIBLE TO COMPLY WITH A REQUEST AFTER BEING OBLIGED WITH A CONCESSION?

“the old give and take… and take”

— Robert Cialdini

Master thesis, specialization Marketing Management

University of Groningen Faculty of Economics and Business

Word count: 6343

June 15th, 2017

Cas de Carpentier Wolf The Hague 0646171413

Student number S1937529 c.a.s.de.carpentier.wolf@student.rug.nl

Supervisor:

Prof. Dr. B. M. Fennis

(3)

ABSTRACT

Stress, could it be that our daily life hassles interfere with our sense of control to the extent that we become more prone to be influenced? From self-regulatory resource depletion theory, we know which consequences impaired self-control could have on our consumer behavior. But what about our self-control in daily life? By means of an experiment was tested whether participants were more susceptible to comply with a request after being induced with stress and obliged with a concession. While the level of experienced stress did not seem to increase the participants’ susceptibility to comply with a request after exposure to a reciprocity appeal - even after controlling for pre-existing stress coping strategies resembling in the participants - the main effects of the reciprocity appeal did seem to significantly increase compliance rates. These results, firstly, reaffirm the strength of the persuasion technique of reciprocation. Secondly, rule out the influence of social stress on compliance behavior, thereby paving the way for other forms of stress to be investigated. And lastly, provide insights into personal traits and circumstances influencing the level of experienced stress.

Keywords: reciprocity, stress, self-control, social influence, stress coping, life history theory

strategies, socioeconomic status, dispositional negativity

(4)

TABLE OF CONTENTS

INTRODUCTION ... 1

THEORETICAL FRAMEWORK ... 3

Social influence, self-control and compliance ... 3

Stress and coping styles ... 4

The principle of reciprocity ... 6

Conceptual framework ... 6

METHODOLOGY ... 7

Participants and Design ... 7

Procedure ... 7

Operationalization Independent Variables ... 8

Operationalization Dependent Variables ... 9

Control Variables ... 11

RESULTS AND DISCUSSION ... 12

Manipulation checks ... 12

Stress manipulation check ‘current stress level’ ... 12

Stress manipulation check ‘stress task assessment’ ... 12

Dependent variables ... 13

Main dependent variable - compliance (amount of time willing to help) ... 13

Additional dependent variable ‘number of acquaintances’ ... 14

Control Variables ... 15

GENERAL DISCUSSION ... 18

LIMITATIONS AND FUTURE RESEARCH DIRECTIONS ... 19

REFERENCES ... 20

APPENDIX ... 24

(5)

INTRODUCTION

How stressed are you at this moment? You might have just received a notification waiting for your direct response. It could be that you have a job interview coming up. Or it might be that the deadline is approaching of a project that is just not getting along. All of these situations, expected or unexpected, might challenge your ability to cope with them, and therefore might result in you experiencing stress (Lazarus, 1966).

While stress may derive from multiple sources, it also tends to lead to different kinds of psychological, physiological and behavioral reactions (Starcke & Brand, 2016). Depending on personal differences, the extent to which people adapt their behavior to a stressful situation may differ substantially across individuals (Kudielka, Hellhammer & Wust, 2009). However, one of the psychological reactions to stress that presumably holds for all of us, is that the adaption of our behavior to stress, hence, our coping with stress, requires the exertion of self- control (Muraven & Baumeister, 2000).

The exertion of self-control requires strength. Similar processes such as volition, turning down impulses and resisting forms of influence, all require strength from resources with a limited capacity (Fennis & Stroebe, 2016). Moreover, when we direct our self-control efforts towards one particular goal (e.g. stress coping), the amount of strength available for a subsequent effort is reduced (Muraven & Baumeister, 2000).

This psychological mechanism of self-regulatory resource depletion (Baumeister, Bratslavsky, Muraven & Tice, 1998) and more specifically, the state in which our resources are depleted, has considerable consequences for our consumer behavior. It increases our propensity to rely on simple heuristics in decision making, thereby increasing susceptibility to comply with requests (Fennis, Janssen & Vohs, 2009), it make us more prone to social influence techniques (Vohs, Fennis & Janssen, 2009) and it may lead to heightened impulse spending tendencies (Vohs & Faber, 2007).

Coming back to the aforementioned situations, what about self-control in daily life?

Could it be that demands of daily life hassles are exceeding our capacity to handle on a day- to-day basis? Are we more prone to be influenced when we experience stress? And is this effect similar for everyone or does it depend on personal stress coping styles? In order to answer these questions, the aim of the present research is to empirically investigate whether people’s susceptibility to comply with a request, after being exposed to the forces of social influence, is higher when they experience stress.

The answer to this question could have significant practical implications for the field

of marketing practice in general and consumer behavior in particular. Not only will it

(6)

contribute to the knowledge of companies and their ability to create more effective advertising strategies. Also for consumers, it may provide insights into a better understanding of their own behavior when experiencing stress.

Furthermore, it contributes to the academic field of marketing research in which the concept of stress is at its early stage. In the recent work of Durante & Laran (2016), results proved that stress does influence the way consumers allocate their money. Showing either increased saving or increased spending behavior depending on the way products are perceived as necessities. But whether stress influences our compliance with a request through social influence is yet to be found out.

Indeed, research proved that when our resources of self-control are impaired, we become more susceptible to influence (Burkley, 2008). Other valuable insights were gained by the work of Fennis et al. (2009) which provided a better understanding of how to induce this state of resource depletion, and how this process is a key reason for social influence tactics to be effective. The present study would be the first to empirically investigate whether the effects of stress from daily life hassles could be comparable with the effects of depleted resources, in that these effects might increase susceptibility to comply with a request.

Outline. In the following chapter, an elaboration on the research question will continue, providing an extended theoretical context and corresponding hypothesis. The second chapter encloses the methods used in testing the hypothesis and the operationalization of the variables. Concluded will be with the results of the experiment, a discussion and implications section including limitations and future research.

(7)

THEORETICAL FRAMEWORK Social influence, self-control, and compliance

Social influence as addressed in the present article fits in the field of Personal Selling as a component of the integrated marketing communication mix, see Figure 1 for an overview.

Traditional Marketing mix

Promotion mix Integrated marketing

communications

Product Direct (response)

marketing

Price Interactive marketing

Marketing

Promotion Sales promotion

Place Public relations

Personal Selling

Fig. 1: Overview scope of present research

Compared to the influence principles in conventional advertising, this form of

communication is characterized by two-way, face-to-face interactions to inform and persuade

consumers, with the intention to yield a behavioral response (Fennis & Stroebe, 2016). The

consumer is exposed to a request, often through sophisticated persuasion strategies (Cialdini,

2009), in their actual decision environment. For instance, at the point-of-purchase inside a

store where products are bought, in public places on the street where influence agents are

(8)

asking consumers to donate money to charity or in any other form of getting a target to say

‘yes’ to a particular request of compliance.

To better understand the subject compliance is to better understand the psychological phenomena that are at play when we, human beings, respond to social influence forces. From the psychological theory of Freud, dating as far back as 1923, we know that volition is one of the self’s crucial functions (Freud, 1923/1961). Hence, controlling the environment, controlling the self, making choices and initiating action. However, as Baumeister, Bratslavsky, Muraven and Tice (1998) discovered, the resources for exerting active volition, hence regulation of the self, are limited. So, whenever we make choices or decisions, take responsibility, override initial responses, hence any kind of deliberate, conscious and controlled behavior, we consume strength from internal self-control resources with a finite capacity (Muraven, Tice & Baumeister, 1998). Thereby temporarily exhausting this resource, leaving less self-regulation strength for any subsequent effort of self-control (Baumeister et al., 1998).

The link between the self-regulatory resource depletion theory, social influence techniques, and compliance was constructed by the work of Fennis et al. (2009) and their developed two-stage model. In the first stage, the consumer is exposed to an initial request.

Consciously attending and responding to this request produces self-regulatory resource depletion. In the second stage, the consumer is in a weakened volitional state, referred to by Fennis et al. (2009) as depletion-induced mindlessness. This state of mindlessness affects compliance through reliance on heuristics and hence, encourages yielding to the target request.

From this line of research could be inferred that when consumers are indeed affected by the demands of a stressful situation and draw on their self-control resources, this depletion would affect their decision making and presumably their susceptibility to comply with a request trough exposure to a heuristic cue. Hence, the following hypothesis is proposed:

H1: The effect of a reciprocity appeal on the compliance behavior of consumers is moderated by their level of experienced stress.

Stress and coping styles

Stress, as addressed in the present article, is defined as “an event or situation that is

evaluated by the consumer as stressful and therefore requires mental and behavioral

adjustments within a relatively short period of time” (Moschis, 2009, p. 431). Furthermore, it

is characterized by a social component. Described by Dickerson and Kemeny (2004) as a

(9)

social-evaluative threat, occurring when an aspect of the self-identity is or could be negatively judged by others. Tasks as public speaking and anticipation on evaluative judgment evoke this kind of social stress response (Kirschbaum, Pirke & Hellhammer, 1993).

As the mental and behavioral adjustment caused by the social stress response of the consumer are assumed to lead to depletion of self-control, and therefore to higher susceptibility to compliance the next hypothesis will be examined:

H2: Consumers experiencing stress will be more likely to comply with a request than consumers experiencing no stress.

Life History Strategies and Socioeconomic Status. Depending on their childhood environment, people might respond differently to current environmental uncertainties (Griskevicius & Kenrick, 2013). According to the life history theory (Kaplan & Gangestad, 2005) the nature and evolution of the environment an individual had, results in people adopting either a fast or a slow life history strategy. Meaning that individuals allocate their limited resources differently. These individual differences develop partly in response to early life experiences (Mittal & Griskevicius, 2014). For example, a sense of current environmental uncertainty may lead to slower strategy responses (i.e. becoming less impulsive, taking fewer risks and delaying gratification) for people coming from wealthier childhood environments.

While on the other hand, people with relatively poor childhood backgrounds may show the opposite pattern in a faster strategy (i.e. becoming more impulsive, taking more risk and seeking immediate gratification) to the same uncertainty.

Dispositional negativity. Dispositional negativity is a personal trait that is expressed by the tendency to perceive environmental phenomena with a more negative affectivity (Shackman et al., 2016). A heightened negative affect or individuals with a more negative disposition tend to have a higher negative affect towards stressors, might more often experience stress even in relaxed and familiar situations and tend to make the stressful situation even more stressful by acting in ways that increases the likelihood of conflict (Shackman et al., 2016).

From this line of research, it may be inferred that although an event is evaluated by one consumer as stressful (with adjustment of behavior and hence, self-control depletion as a consequence) while another consumer might keep his cool and does not draw from his regulatory resources and hence, not change his compliance behavior. Hence, the hypothesis:

H3: The influence of ‘life history theory strategies’, ‘socioeconomic status’, and

‘dispositional negativity’ will affect the level of experienced stress of consumers.

(10)

The principle of reciprocity

To get a better grasp of the heuristic referred to in the preceding line of reasoning, the theoretical framework will be concluded with a paragraph ascribed to the principle of reciprocity. Reciprocation basically says, “you should try to repay, in kind, what another person has provided you” (Cialdini, 2009, p. 13). Gouldner (1960) referred to reciprocation as

“the essence of what makes us human” and if it was not for this simple rule of reciprocity our ancestors would not have learned to share their food and their skills in an honored network of obligation. As such it is a principle very deeply engrained in the human psyche.

The door-in-the-face technique is one of the classic examples of how an unsolicited obligation affects compliance with a request (Cialdini et al., 1975). Starting with a relatively large request, most likely to be rejected, the next request (the actual target request) is more moderate. By sizing down the initial request, a clear unsolicited concession is made, most evoking the need to make a concession in return by the consumer by complying with the moderate request. This brings us to the final hypothesis:

H4: Exposing consumers to a reciprocity appeal increases their susceptibility to comply with a request.

Conceptual framework

A depiction of the preceding reasoning and assumed relations between the discussed concepts is proposed in the following conceptual framework, see Figure 2.

Fig.2: Conceptual framework

Social stress Stress condition vs.

no stress condition Reciprocity

Reciprocity appeal vs.

no reciprocity appeal

Compliance Amount of time willing

to help

Stress Coping - Life History Theory - Socioeconomic Status - Dispositional Negativity H4

H2

H1

H3

(11)

METHODOLOGY Participants and Design

A total of 118 Dutch citizens (51 males, 67 females; M age = 31.88 years; SD = 15.46) participated in a field experiment on a voluntary basis. Two participants did not pass the two control questions (asking whether they were actually reading the questions) and were excluded from further analysis. Participants were approached in public places and the college building of The Hague University of Applied Sciences, and selected based on three criteria: 1) minimum estimated age of 18 years; 2) behavior that indicated they had the time to participate in the experiment (e.g. sitting, presumably waiting); and 3) master the Dutch language.

Participants did not have any prior knowledge about the study. The only information they received was that the research was executed in the context of a Master Thesis for the University of Groningen.

The study design was a 2 (no stress (control) condition vs. stress condition) x 2 (no reciprocity appeal (control) vs. reciprocity appeal) between-subjects factorial design.

No stress condition Stress condition

No-reciprocity appeal Condition 1 (no reciprocity/no stress)

Condition 2 (no reciprocity/stress)

Reciprocity appeal Condition 3

(reciprocity/no stress)

Condition 4 (reciprocity/stress) Table 1: Study Design

Procedure

Participants were told that the purpose of the research was to examine the potential difference between people from the north and the west of the Netherlands in how they process different kinds of information, as adapted and modified from Durante and Laran (2016).

Participants were randomly allocated to one of the four experimental conditions by receiving one of the four different versions of the hardcopy questionnaire, see Appendix B for version 1 and 4.

On the first page of the questionnaire participants were asked for their demographic

information. The next part comprised the manipulation of the stressors, consisting of a stress

task for participants in the stress condition and a no stress task for participants in the no stress

condition. The purpose of this manipulation was to induce either a stress response or a no

stress response (i.e. control condition).

(12)

The following section involved the manipulation of the salience of the reciprocity appeal. By making a concession to the participants in the reciprocity appeal condition, the aim of the manipulation was to activate the mechanism of concession in return in the form of increased compliance with a request of compliance (Fennis et al., 2009) as measured in the subsequent section. Participants in the no reciprocity condition were not made a concession and completed the questionnaire without interruption of the experimenter.

Following the reciprocity manipulation, participant’s compliance was measured by indicating the amount of time they were prepared to volunteer in a fictional follow-up study (Kardes, Fennis, Hirt, Tormala & Bullington, 2007). Directly thereafter followed two stress manipulation checks. Participants were asked to answer questions about how they were feeling in that moment of time (Durante & Laran, 2016) and about how they experienced the stress manipulation task (Acar-Burkay, Fennis & Warlop, 2014).

In the last part of the questionnaire, participants were asked to fill in three scales indicating personal traits and their environmental circumstances. After completion of the questionnaire, participants were debriefed and heartily thanked.

Operationalization Independent Variables

Stress response. The level of experienced stress was manipulated by a modification of the Trier Social Stress Test (TSST; Kirschbaum et al., 1993). The TSST is a procedure used to reliably induce a stress response in research participants (Kirschbaum et al., 1993). It consists of different stress generating tasks (i.e. public speaking, mental arithmetic calculations, and anticipation). Dickerson and Kemeny (2004) considered the TSST to be the best available procedure to elicit psychological stress responses. An important aspect of the TSST is the socially evaluative component, described by Dickerson and Kemeny (2004) as the social-evaluative threat.

In the present experiment, participants in the stress condition were asked to prepare a 30-second presentation to be presented at the end of the survey. They were given two minutes to prepare and suggested to take notes. To strengthen the social-evaluative threat component, the topic of the presentation was a personal trait which the participant wanted to change.

Moreover, it was pointed out that after the presentation the participant was to be asked two

substantive questions and that the overall performance was to be graded and compared with

the performance of other participants. After this anticipatory stress phase (Kirschbaum et al.,

1993) followed the mental arithmetic questions as a set-up for the reciprocity appeal.

(13)

Participants in the no stress condition were not asked to prepare a presentation, but to describe a personal trait which they wanted to change.

Reciprocity appeal. The operationalization of the reciprocity appeal was adapted from a previous study of Fennis et al. (2009). The aim of the manipulation of the reciprocity appeal was to, firstly let the experimenter make a concession to the participant, so consequently, this would induce a counter-concession on the part of the participant (Cialdini, 2009). The concession made by the experimenter consisted of making an exception and excusing the participant from filling out a page with rather difficult arithmetic questions. The expected counter concession made by the participant consisted of increased compliance with the volunteering request.

Participants in the reciprocity appeal condition were asked to answer ten arithmetic questions, Appendix A.1. However, when the participants were about to begin with the first question, the experimenter interrupted and excused the participant of actually filling out these arithmetic questions: “Since I have collected enough data on these questions and because they were described by other participants as rather difficult, I will make an exception for you.

You may skip this part and continue with the next section.” The questionnaires of the participants in the no-reciprocity condition did not contain arithmetic questions. Hence, these participants were not excused of anything and completed to questionnaire without interruption of the experimenter.

Results of the study of Fennis et al. (2009) showed reliable differences in compliance between the reciprocity appeal condition versus the no-reciprocity condition.

Operationalization Dependent Variables

Compliance. Following a procedure used by Kardes et al. (2007) and Fennis et al.

(2009) the dependent variable of compliance was measured as the amount of time the participant would volunteer in a fictive online follow-up study of the experimenter.

Participants were told that the present questionnaire was only the first stage of the research of the experimenter. For the second stage, he would again rely on the scarce availability of participants. Therefore, participants were asked if they would do the experimenter a favor by volunteering in his follow-up study. Participants could answer this target request on a scale ranging from 0 to 90 minutes in 10-minute intervals as adapted from Fennis et al. (2009). The amount of time participants were willing to help served as the measure of compliance.

Subsequently, participants were asked to fill in their mail address. Thereby making the

scenario more realistic and support the intention of the indicated length of time.

(14)

In addition, a second measure of compliance was administered by asking the participant to fill in names and mail addresses of acquaintances who might be willing to volunteer in the follow-up study. Hence, it was registered if, and if so, how many names were provided. This technique is described by Cialdini (2009) as a component of the sales pitch of door-to-door sales operations. In the case of a customer refusing the offer of the salesmen, he would retreat to a request for referrals. Obviously, the main goal of these salesmen is to make the deal, however, a worthy second place achievement could be attributed to getting referrals.

As these prospects might be more willing to buy when the sales operator is able to mention the name of the familiar person who “recommended” the sales visit (Cialdini, 2009).

Although this rejection-then-retreat technique depends on the refusal of the first request, the actual offer, in the present study this second request is measured independent of the compliance of the target request (amount of time willing to volunteer). Expectations are that when the participant indicated not to be willing to help (i.e. indicating 0 minutes), he or she would be more likely to provide a name of an acquaintance.

Stress manipulation checks. Because the stressor manipulation and intended increase in the level of stress were the main issues of the experiment, two measurements of the stress level were administered. Firstly, the “current stress level” scale, serving as a measure of how stressed participants were feeling at the moment in time. Thereby not only accounting for stress experienced by the stress task, but also for other influences of stress, hence potential stress from daily life hassles. The second manipulation check consisted of the “stress task assessment” scale, which served as a measure of how challenging participants experienced the stress task to be.

The former scale was adapted from previous research (Durante & Laran, 2016;

Lovibond & Lovibond, 1995). Participants indicated on a 9-point scale (1 = “Definitely disagree,” and 9 = “Definitely agree”), how stressed they were feeling at that moment in time (e.g. (1) “I find it hard to wind down”, (7) “I am in a state of nervous tension”) see Appendix A.2 for all items. These 8 statements are a sub-scale (measuring the stress dimension) of the Depression Anxiety Stress Scale (Lovibond & Lovibond, 1995). The internal consistency of the scale was found to be highly reliable (8 items; α = .90). A composite score was computed for each participant by averaging the item scores, with higher scores indicating higher levels of stress.

The second scale, the “stress task assessment” was adapted from the research of Acar-

Burkay et al. (2014). On a 7-point scale, participants indicated on six items how demanding

they found the stress manipulation task (e.g. “I found the task”: (1) “difficult”, (2) “stressful”,

(15)

(5) demanding). Three items (“I found the task”: (3) “easy”, (4) “fun”, and (6) “enjoyable”) were reverse coded. The mean scores of the six items including the three reverse coded items were calculated by averaging the item scores, with higher scores indicating that the participant experienced the stress manipulation task to be more demanding and presumably more stressful (6 items; α = .76).

Control Variables

Life History Strategies and Socioeconomic Status. To assess the life history theory strategies, the Mini-K Short Form (Figueredo et al., 2006) was used. This 20-item scale is a component of the Arizona Life History Battery (ALHB; Figueredo, 2007). The ALHB is a measurement of cognitive and behavioral indicators of life history strategy (Olderbak, Gladden, Wolf &

Figueredo, 2013).

Participants were presented 20 statements and indicated to what extent they agreed or disagreed with these on a 7-point scale, see Appendix A.3 for all items. Initially, the internal consistency of these 20 items proved to be rather low (α = .58). An inspection of the analysis indicated that the scale reliability could be improved by eliminating item 6 (“I avoid taking risks”) resulting in a more reliable scale (19 items; α = .61). It seemed that the interpretation of the Dutch translation of this item was ambiguous and therefore it was left out of further analysis. Composite scores were computed by averaging responses to all items, with high scores indicating slow-, and low scores indicating fast life history strategies.

The childhood socioeconomic status (SES) scale comprised 6 items, measured on a 9- point scale as adapted from previous research (Chen & Miller, 2012; Griskevicius, Tybur, Delton & Robertson, 2011) see Appendix A.4 for all items. Mean scores were calculated by averaging the item scores, with high scores indicating a more pleasant and predictable childhood environment and low scores indicating a more harsh and unpredictable childhood environment (Brady & Matthews, 2002) (6 items; α = .82).

Dispositional Negativity. To assess the dispositional negativity traits, the Dispositional Attitude Measure (DAM) (Hepler & Albarracin, 2013) scale was used. This scale consists of 16 words that at first glance might give the impression that they were chosen at random.

However, by asking participants to indicate on a 7-point scale (1 = extremely unfavorable, 7 =

extremely favorable), how they generally feel about topics as (1) architecture, (7) Japan, (10)

public speaking or (14) statistics (see Appendix A.5 for all items), participants’ tendency to

generally like or dislike stimuli, hence, their dispositional attitude could be examined (Hepler

(16)

& Albarracin, 2013). The mean scores were calculated by averaging the scores of all items, with low scores indicating an overall tendency to have a more negative attitude regardless of what stimuli are being evaluated then participants with higher scores, and vice-versa. (16 items; α = .71).

RESULTS AND DISCUSSION Manipulation checks

Stress manipulation check ‘current stress level’

A first analysis was conducted in order to check whether the stress manipulation succeeded in forming higher stress levels for the participants in the stress conditions. Hence, a 2 (no stress condition vs. stress condition) x 2 (no reciprocity appeal vs. reciprocity appeal) ANOVA was conducted on the measure of the ‘current stress level’.

The stress manipulation proved no significant main effect the ‘current stress level’, (F(1, 112) = 0.050, p = .823, partial η2 = .000). Despite the non-significance, against all odds, the ‘current stress level’ was marginally higher for the ‘no stress condition’ (M = 28.683, SD

= 12.817) then for the ‘stress condition’ (M = 28.143, SD = 13.449). Also the main effect of reciprocity was not significant (F(1, 112) = .019, p = .891, partial η2 = .000). However, this latter result is in line with expectations, since the effect of reciprocity was not expected to produce any change in experienced stress.

Somewhat notable was the significant interaction effect between stress and reciprocity on ‘current stress level’ scores (F(1, 112) = 3.998, p = .048, partial η2 = .034). Suggesting that the effect of the stress manipulation on the ‘current stress level’ depended on the reciprocity manipulation.

Stress manipulation check ‘stress task assessment’

The second check of the stress manipulation was conducted by performing a 2 (no stress condition vs. stress condition) x 2 (no reciprocity appeal vs. reciprocity appeal) ANOVA on the self-reported ‘assessment of the stress manipulation task’.

For this second stress manipulation check the stress manipulation did prove to be successful as it showed a significant main effect (F(1, 112) = 4.727, p = .032, partial η2

= .040). In line with our expectations the ‘stress condition’ (M = 20.250, SD = 5.785) resulted

in higher scores for the ‘assessment of the stress manipulation task’ than for the ‘no stress

condition’ (M = 18.050, SD = 5.150). Suggesting that participants in the stress condition did

experience higher stress levels or experienced the stress task to be more of a challenge than

participants in the ‘no stress condition’.

(17)

In line to the previous (first) stress manipulation check and our expectations, the main effect of reciprocity was not statistically significant (F(1, 112) = 1.818, p = .180, partial η2

= .016). Furthermore, the results showed no significant interaction effect between stress and reciprocity on the self-reported ‘assessment of the stress manipulation task’ (F(1, 112) = 1.106, p = .295, partial η2 = .010).

Dependent variables

Main dependent variable - compliance (amount of time willing to help)

To test our main hypothesis that people’s susceptibility to comply with a request after being persuaded by the marketing technique of reciprocity is higher when they experience stress, a 2 (no stress (control) condition vs. stress condition) x 2 (no reciprocity appeal (control) vs. reciprocity appeal) ANOVA on the ‘amount of time willing to help’ was conducted.

Residual analysis was performed to test the assumptions of the two-way ANOVA analysis procedure. Since the assumptions of no outliers and normality seemed to be violated for the “amount of time willing to help scores, the next step in the analysis consisted of running test comparisons with transformed data in order to check whether the outliers and the lack of normality were affecting the results in a meaningful way, see Appendix C.1 for results. Eventually, by comparing the results of two-way ANOVA of the original (non- transformed) data and the results of the two-way ANOVA of the transformed data, the statistical conclusions did not seem to be appreciably different (see Appendix C.2) and therefore, we continued our analysis with the original non-transformed data.

The interaction effect between stress and reciprocity on ‘amount of time willing to help’ was not significant (F(1, 112) = .204, p = .652, partial η2 = .002). Therefore, an analysis of the main effects was performed. The main effect of stress was not significant (F(1, 112) = 0.087, p = .769, partial η2 = .001). However, despite the non-significance, the mean scores of the stress manipulation resulted in contradicting our hypothesis: with a marginally higher compliance rate for the ‘no stress condition’ (M = 1.500, SD = 1.372) compared to the compliance rate of the ‘stress condition’ (M = 1.429, SD = 1.346).

The reciprocity manipulation did prove to show a statistically significant main effect (F(1, 112) = 11.557, p = .001, partial η2 = .094). In line with the hypothesis and previous findings the ‘reciprocity condition’ (M = 1.879, SD = 1.534) showed a significant higher mean score of compliance compared to the ‘no reciprocity condition’ (M = 1.0517, SD

= .999).

(18)

Diving deeper in the details, it seemed that compared to the control condition (condition 1; no reciprocity/no stress) (M = 1.033, SD = 1.033), the highest mean score of compliance was achieved in the reciprocity/no stress condition (condition 3) (M = 1.967, SD

= 1.520). Followed by the second highest compliance mean score achieved in the reciprocity/stress condition (condition 4) (M = 1.786, SD = 1.572). Reaffirming that the reciprocity appeal was successful.

Pairwise comparisons were run of where reported 95% confidence intervals and p- values are Bonferroni-adjusted. The unweighted marginal means of ‘level of compliance’

scores were (M = 1.052, SE = 0.171) for the no-reciprocity condition and (M = 1.876, SE = 0.171) for the reciprocity condition. The reciprocity condition was associated with a mean

“amount of time willing to help’’ score (M = .824, 95% CI [.344, 1.304] higher than the no- reciprocity condition with a statistically significant difference (p = .001).

Additional dependent variable ‘number of acquaintances’

A total of 14 participants provided the name and mail address of acquaintances whom might be willing to help. As hypothesized based on the rejection-then-retreat technique (Cialdini, 2009), participants who were not willing to help, hence indicating ‘0 minutes’ were expected to be more likely to provide names of acquaintances. However, of those 14 participants who provided at least 1 name, only 3 participants indicated 0 minutes.

Notably, despite contradicting the expectations based on the rejection-then-retreat technique, the highest mean was achieved in condition 4 (stress/reciprocity) (M = .321, SD

= .723). Suggesting that, participants did seem to be more prone to comply with a request for acquaintances, when under stress and persuaded by the reciprocity appeal.

No stress condition Stress condition Total

No-reciprocity appeal

Condition 1 (no-reciprocity/no stress)

(M = .133, SD = .346)

Condition 2 (no-reciprocity/stress) (M = .000, SD = .000)

n = 58 (M = .069, SD = .256) Reciprocity appeal

Condition 3 (reciprocity/no stress) (M = .233, SD = .774)

Condition 4 (reciprocity/stress) (M = .321, SD = .723)

n = 58 (M = .276, SD = .744)

Total n = 60

(M = .183, SD = .596)

n = 56

(M = .161, SD = .532) n = 116 Table 2: Means and Standard deviations for additional dependent variable ‘number of acquaintances’

(19)

Control Variables

In order to statistically control for pre-existing differences between the treatment groups due to the confounding (control) variables, life history strategy, socioeconomic status, and dispositional negativity traits, potentially affecting the dependent variable compliance;

three separate ANCOVAs were conducted to determine the effect of the stress manipulation (stress and no stress) on “the amount of time willing to help” scores after controlling for “life history strategy”, “socioeconomic status” and “dispositional negativity”. As hypothesized these variables might affect the way in which a person reacts to a particular uncertainty in the environment (i.e. our stress manipulation task).

Since the measurement of the covariates occurred after the stress and reciprocity manipulations, this sequence might have affected the scores of our covariates. Therefore, we conducted three separate 2 (no stress condition vs. stress condition) x 2 (no reciprocity appeal vs. reciprocity appeal) ANOVAs on all three covariate factors as a dependent variable. Since the stress and reciprocity manipulations did not affect the covariate scores significantly we persecuted with testing for the assumption of homogeneity of regression. There were no significant findings for any of the interactions. Hence, we assumed to have met the assumption of homogeneity of regression. However, interpretation of the results was done with caution since the assumption of linearity in the relationship between the compliance scores and covariate scores was violated.

In line with the previous two-way ANOVA analysis on the dependent variable

compliance, the three separate ANCOVA analyses resulted in a significant main effect of the

reciprocity appeal, no significant main effects of stress, no interaction effects of stress and

reciprocity nor any significant findings for the covariates, table 3. Suggesting that the

covariates did not significantly predict the ‘amount of time willing to help’. However, from

these three covariates the “life history strategy” seemed to have reduced the most unexplained

variance (error term) with 3.631 units compared to 1.328 units for the dispositional negativity

and .496 units for the socioeconomic status, respectively.

(20)

Control variables Source F Sig. Partial Eta Squared

Life History Strategies

Covariate LHT F (1,111) = 2.157 p = .145 η2 = .019 Main effect stress F (1,111) = .108 p = .744 η2 = .001 Main effect reciprocity F (1,111) =

13.002** p = .000** η2 = .105**

Interaction effect stress*reciprocity F (1,111) = .197 p = .658 η2 = .002

Socioeconomic Status

Covariate SES F (1,111) = .290 p = .591 η2 = .003 Main effect stress F (1,111) = .052 p = .820 η2 = .000 Main effect reciprocity F (1,111) =

11.233** p = .001** η2 = .092**

Interaction effect stress*reciprocity F (1,111) = .131 p = .718 η2 = .001

Dispositional Negativity

Covariate DAM F (1,111) = .779 p = .379 η2 = .007 Main effect stress F (1,111) = .132 p = .717 η2 = .001 Main effect reciprocity F (1,111) =

11.925** p = .001** η2 = .097**

Interaction effect stress*reciprocity F (1,111) = .431 p = .513 η2 = .004 Table 3: DV: compliance; Results ANCOVAs per covariate

After ruling out the potential influence of the covariates on the dependent variable compliance and in order to test the hypotheses of the covariates affecting the way a person reacts differently to uncertainties, a series of ANCOVAs was run to determine the effects of the stress manipulation (stress vs. no stress) on the first and the second stress manipulation checks: “current stress level” and ‘stress task assessment’, after controlling for “life history strategy”, “socioeconomic status” and “dispositional negativity” traits, see Table 3.

First stress manipulation check. For the first stress measurement “current stress level”, which not only accounted for stress experienced by the stress task, but also for other influences of stress, hence potential stress from daily life hassles, the covariate “life history strategy” proved to predict the scores significantly (F(1, 113) = 13.376, p = .000). The other covariates “socioeconomic status” and “dispositional negativity” did not seem to predict the scores of the “current stress level” significantly. Furthermore, it seems that even when controlling for the three covariates, the stress manipulation did not seem to affect the “current stress level” scores significantly.

Second stress manipulation check. For the second stress measurement “stress task

assessment” the covariate “dispositional negativity” did prove to predict to “stress task

(21)

assessment” scores significantly (F(1, 113) = 7.496, p = .007). The covariates “life history theory” and “socioeconomic status” did not prove to predict the scores significantly, table 4.

Notably, even when controlling for the three covariates, the “stress task assessment” scores did prove to differ for our stress manipulation with higher mean scores for the “stress condition” compared to the “no stress condition”.

Control

variable Dependent variable Source F Sig. Partial Eta

Squared

Life History Strategies

First stress manipulation check

“current stress level”

(Levene’s test (p = .675)

Covariate LHT**

F (1,113) =

13.376** p = .000** η2 = .106**

Main effect stress

F (1,113)

= .024 p = .877 η2 = .000

Second stress manipulation check

“stress task assessment”

(Levene’s test (p = .341)

Covariate LHT F (1,113)

= .143 p = .706 η2 = .001 Main effect

stress*

F (1,113) =

4.691* p = .032* η2 = .040*

Socioeconomic Status

First stress manipulation check

“current stress level”

(Levene’s test (p = .792)

Covariate SES F (1,113) =

1.253 p = .265 η2 = .011 Main effect

stress

F (1,113)

= .008 p = .929 η2 = .000

Second stress manipulation check

“stress task assessment”

(Levene’s test (p = .355)

Covariate SES F (1,113)

= .816 p = .368 η2 = .007 Main effect

stress*

F (1,113) =

5.068* p = .026* η2 = .043*

Dispositional Negativity

First stress manipulation check

“current stress level”

(Levene’s test (p = .878)

Covariate DAM F (1,113)

= .216 p = .610 η2 = .002 Main effect

stress

F (1,113)

= .033 p = .856 η2 = .000

Second stress manipulation check

“assessment stress task”

(Levene’s test (p = .336)

Covariate DAM*

F (1,113) =

7.496* p = .007* η2 = .062*

Main effect stress*

F (1,113) =

5.902* p = .017* η2 = .050*

Table 4: DV’s: “current stress level” & “assessment stress task”; Results ANCOVAs per covariate

(22)

GENERAL DISCUSSION

The aim of this research was to investigate whether people’s susceptibility to comply with a request after being persuaded by the marketing technique of reciprocity is higher when they experience stress. It was assumed that by increasing the level of experienced stress, a person is more likely to comply with a request of helping out in a (fictitious) follow-up study when they were exposed to a concession made by the experimenter. We tested this notion by means of an experiment in which both the level of stress and a reciprocity appeal were manipulated.

Stress manipulation. In order to check whether the stress manipulation was successful in increasing the level of experienced stress of the participants, two stress manipulation checks were assessed. One to assess the ‘current stress level’ (Durante & Laran, 2016) and a second one to assess the way participants experienced the stress manipulation task to be demanding.

For the first manipulation check the stress manipulation did not have a significant effect on the ‘current stress level’. It even contradicted our expectations with higher levels of

‘current stress’ for the no stress conditions than for the stress conditions. It seemed that the

‘current stress level’ was subject to influences other than the type of condition (e.g. an exam or an approaching deadline). The second manipulation check off the stress manipulation did prove to have a significant effect on the way participants experienced the stress manipulation task to be challenging/stressful. These findings extend the empirical employability of the Trier Social Stress Test (Kirschbaum et al., 1993), from being limited to a laboratory setting only, to applicability for non-laboratory (natural field research) settings. Although this procedure was modified for the present research it did prove to be successful to a certain extent.

Compliance. While the stress manipulation did prove to be successful for at least one of the manipulation checks, it did not increase the compliance rate significantly. This main finding contributes to the general knowledge in the field of consumer behavior in which the effects of the construct stress is at its early stage. For the reciprocity appeal, we did see a significant increase in the compliance rate. This finding reaffirms the strength of the good old principle of reciprocity.

Covariates. Although participants were randomly assigned to a condition, this could have led to values of the group covariates means to be unequal. Controlling for the effects of the covariates “life history strategies”, “socioeconomic status” and “dispositional negativity”

traits, did not significantly affect the results of our compliance measure. However, they did

(23)

seem to have some influence. This means that these variables cannot be used to explain differences

LIMITATIONS AND FUTURE RESEARCH DIRECTIONS

A few limitations of the present research should be noted. As the stress manipulation and subsequent measurement were an important aspect in the present study, two measurements of experienced stress were implemented. Although both measurements seemed to reliably measure the underlying constructs “current stress level” and “challenging nature of the stress task”, their results differed when comparing the same conditions of the study design. Hence, in a laboratory experiment a more controlled and objective approach to measure the stress level of participants could be initiated by measuring the level of the cortisol hormone (Dickerson & Kemeny, 2004).

Second, the fact that the participants were approached in a college building might have been of influence on the results of the compliance measure since students among students might be more willing to help out each other. Therefore, higher scores of compliance might have been indicated compared to when participants were exclusively approached in public places. However, student or no student, volunteering time should be independent of a person’s occupancy. Another way for future studies to measure compliance could be to actually ask for a (small) amount of money (e.g. for donation to charity) thereby measuring actual spending behavior.

Third, the scope of the present study was focused on social stress, however, there are multiple forms stressful situations applicable for experimental research. Perceived lack of money stress might induce stress, where it might by that people are more willing to help out.

Furthermore, it could be tested whether an extension of the favor, hence making a larger favor to the participant (e.g. giving a voucher), would further increase compliance rates. Lastly, the experiments could be conducted with different experimenters, thereby ruling out potential influence if another well-known rule of thumb, the heuristic ‘liking’ (Fennis & Stroebe, 2016).

In sum, it can be stated that the expected relation between stress and the susceptibility

to the marketing technique of reciprocity is not as simple as expected. The level of

experienced stress appears to not directly increase people’s willingness to comply with a

certain request after having been exposed to a concession made be an influence agent (the

experimenter). This finding might of particular interest for the confutation of further research

on the influence of stress on consumer behavior.

(24)

REFERENCES

Acar-Burkay, S., Fennis, B., & Warlop, L. (2014). Trusting others: the polarization effects of need for closure. Journal of Personality and Social Psychology, 107 (4), 719-735.

Baumeister, R.E., Bratslavsky, E., Muraven, M., & Tice, D.M. (1998). Ego depletion: is the active self a limited resource? Journal of Personality and Social Psychology, 74 (5), 1252–1265.

Brady, S.S., & Matthews, K.A. (2002). The influence of socioeconomic status and ethnicity on adolescents' exposure to stressful life events. Journal of Pediatric Psychology, 27 (7), 575-583.

Burkley, E. (2008). The Role of Self-Control in Resistance to Persuasion. Personality and Social Psychology Bulletin, 34 (3), 419-431.

Chen, E., & Miller, G.E. (2012). “Shift-and-persist” strategies: why low socioeconomic status isn’t always bad for health. Perspectives on Psychological Science, 7 (2), 135- 158.

Cialdini, R.B., & Goldstein, N.J. (2004). Social Influence: compliance and conformity.

Annual Review of Psychology, 55 (1), 591-621.

Cialdini, R.B. (2009). Influence: The psychology of persuasion (Revised Edition). New York:

Harper Collins

Cialdini, R.B., Vincent, J.E., Lewis, S.K., Catalan, J., Wheeler, D., & Darby, B.L. (1975).

Reciprocal concessions procedure for inducing compliance: the door-in-the-face technique. Journal of Personality and Social Psychology, 31 (2), 206-215.

Dickerson, S.S., & Kemeny, M.E. (2004). Acute stressors and cortisol responses: a

theoretical integration and synthesis of laboratory research. Psychological Bulletin,

130 (3), 355–391.

(25)

Durante, K.M., & Laran, J. (2016). The effect of stress on consumer saving and spending.

Journal of Marketing Research, 53 (5), 814–828.

Fennis, B.M., & Stroebe, W. (2016). The Psychology of Advertising (second edition). New York: Routledge

Fennis, B. M., Janssen, L., & Vohs, K. D. (2009). Acts of benevolence: a limited-resource account of compliance with charitable requests. Journal of Consumer Research, 35 (6), 906-924.

Figueredo, A.J., Wolf, P.S.A., Olderbak, S.G., Gladden, P.R., Fernandes, H.B.F., Wenner, C., Hill, D., Andrzejczak, D.J., Sisco, M.M., Jacobs, W.J., Hohman, Z.J., Sefcek, J.A., Kruger, D., Howrigan, D., MacDonald, K. & Ruston, J.P. (2014). The psychometric assessment of human life history strategy: a meta-analytic construct validation.

Evolutionary Behavioral Sciences, 8 (3), 148-185.

Freud, S. (1961). The ego and the id. In J. Strachey (Eds.) The standard edition of the complete psychological works of Sigmund Freud (Vol. 19, pp. 3 - 66). London:

Hogarth Press. (Original work published 1923)

Gouldner, A.W. (1960). The norm of reciprocity: A preliminary statement. American Sociological Review, (25) 2, 161-178.

Griskevicius, V., & Kenrick, D.T. (2013). Fundamental motives: How evolutionary needs influence consumer behavior. Journal of Consumer Psychology, 23 (3), 372-386.

Griskevicius, V., Tybur, J.M., Delton, A.W., & Robertson, T.E. (2011). The influence of mortality and socioeconomic status on risk and delayed rewards: A life history theory approach. Journal of Personality and Social Psychology, 100, 1015–1026

Hepler, J., & Albarracin, D. (2013). Attitudes without objects: Evidence for a dispositional

attitude, its measurement, and its consequences. Journal of personality and social

psychology, 104 (6), 1060-1076.

(26)

Kaplan, H. S., & Gangestad, S.W. (2005). Life history theory and evolutionary psychology.

The handbook of evolutionary psychology, 68-95.

Kardes, F.R., Fennis, B.M., Hirt, E.R., Tormala, Z.L., Bullington, B. (2007). The role of the need for cognitive closure in the effectiveness of the disrupt-then-reframe influence technique. Journal of Consumer Research, 34, 377-385.

Kudielka, B.M., Hellhammer, D.H., & Wüst, S. (2009). Why do we respond so differently?

Reviewing determinants of human salivary cortisol responses to challenge.

Psychoneuroendocrinology, 34 (1), 2-18.

Kirschbaum, C., Pirke, K.M., & Hellhammer, D.H. (1993). The 'Trier Social Stress Test'--a tool for investigating psychobiological stress responses in a laboratory setting.

Neuropsychobiology, 28 (1-2), 76-81.

Lazarus, R.S., (1966). Psychological Stress and the Coping Process. New York: McGraw- Hill.

Lovibond, P.F., & Lovibond, S.H. (1995). The structure of negative emotional states:

comparison of the depression anxiety stress scales (DASS) with the beck depression and anxiety inventories. Behaviour research and therapy, 33 (3), 335-343.

Mittal, C., & Griskevicius, V. (2014). Sense of control under uncertainty depends on people’s childhood environment: a life history theory approach. Journal of Personality and Social Psychology, 107 (4), 621-637.

Moschis, G.P. (2007). Stress and consumer behavior. Journal of the Academy of Marketing Science, 35, 430-444.

Muraven, M., & Baumeister, R.F. (2000). Self-regulation and depletion of limited resources:

does self-control resemble a muscle? Psychological Bulletin, 126 (2), 247-259.

(27)

Muraven, M., Tice, D. M., & Baumeister, R. F. (1998). Self-control as a limited resource:

Regulatory depletion patterns. Journal of personality and social psychology, 74 (3), 774-789.

Olderbak, S., Gladden, P., Wolf, P.S.A., & Figueredo, A.J. (2013). Comparison of life history strategy measures. Personality and Individual Differences, 58, 82-88.

Shackman, A.J., Tromp, D.P.M., Stockbridge, M.D., Kaplan, C.M., Tillman, R.M., Fox, A.S.

(2016). Dispositional negativity: an integrative psychological and neurobiological perspective. Psychological Bulletin, 142 (12), 1275-1314.

Starcke, K., & Brand, M. (2016). Effects of stress on decisions under uncertainty: A meta- analysis. Psychological Bulletin, 142 (9), 909-933.

Vohs, K.D., & Faber, R.J. (2007). Spent resources: self-regulatory resource availability affects impulse buying. Journal of consumer research, 23, 537-547.

Vohs, K.D., Fennis, B.M., & Janssen, L. (2009). Why do people fall prey to social

influence techniques? A limited-resource account of compliance. In A. L. McGill, &

S. Shavitt (Eds.), Advances in Consumer Research. (Vol. 36, pp. 64-67). Duluth, MN:

Association for consumer research.

(28)

APPENDIX

TABLE OF CONTENTS

APPENDIX A: QUESTIONNAIRE COMPONENTS A.1. Arithmetic questions

A.2. Current stress level A.3 Mini-K Short Form

A.4 Socioeconomic Status Scale A.5 Dispositional Negativity

APPENDIX B: COMPLETE QUESTIONNAIRES

B.1 Questionnaire version 1 - Condition 1: No reciprocity / no stress B.2 Questionnaire version 4 - Condition 4: Reciprocity / stress APPENDIX C: ANALYSIS AND RESULTS

C.1 Residual analysis, results test for assumptions ANOVA

C.2 Re-test transformed data

(29)

APPENDIX A.1: QUESTIONNAIRE COMPONENTS: Arithmetic questions 1. 0.0004 × 3,000 =

2. =

3. In onderstaand figuur, z = 50, wat is de waarde van x + y:

4. 0.00049 : 0.07 = 5. √(16) + (36) =

6. 3 x A = 36. Wat is A:

7. √(16)(20)+(8)(32) =

8. In onderstaand figuur, wat is de waarde van v + x + y + z + w:

9. 9 : 3 + 2 x 4 - 6 =

10. In onderstaand figuur, welk punt heeft de grootste absolute waarde:

(30)

APPENDIX A.2: QUESTIONNAIRE COMPONENTS: Current stress level (1) “I find it hard to wind down,”

(2) “I feel that I am rather touchy.”

(3) “I find myself getting agitated,”

(4) “I have a lot of nervous energy,”

(5) “I find it difficult to relax,”

(6) “I find that I am very irritable,”

(7) “I am in a state of nervous tension,”

(8) “I find myself getting upset,”

APPENDIX A.3: QUESTIONNAIRE COMPONENTS: Mini-K Short Form (1) I can often tell how things will turn out.

(2) I try to understand how I got into a situation to figure out how to handle it.

(3) I often find the bright side to a bad situation.

(4) I don’t give up until I solve my problems.

(5) I often make plans in advance.

(6) I avoid taking risks.

(7) While growing up, I had a close and warm relationship with my biological mother.

(8) While growing up, I had a close and warm relationship with my biological father.

(9) I have a close and warm relationship with my own children.

(10) I have a close and warm romantic relationship with my sexual partner.

(11) I would rather have one than several sexual relationships at a time.

(12) I have to be closely attached to someone before I am comfortable having sex with them.

(13) I am often in social contact with my blood relatives.

(14) I often get emotional support and practical help from my blood relatives.

(15) I often give emotional support and practical help to my blood relatives.

(16) I am often in social contact with my friends.

(17) I often get emotional support and practical help from my friends.

(18) I often give emotional support and practical help to my friends.

(19) I am closely connected to and involved in my community.

(20) I am closely connected to and involved in my religion.

APPENDIX A.4: QUESTIONNAIRE COMPONENTS: SES

(1) My family usually had enough money for things when I was growing up (2) I grew up in a relatively wealthy neighborhood

(3) I felt relatively wealthy compared to the other kids in my school (4) I have enough money to buy things I want

(5) I don’t need to worry too much about paying my bills

(6) I don’t think I’ll have to worry about money too much in the future

(31)

APPENDIX A.5: QUESTIONNAIRE COMPONENTS: Dispositional Negativity (1) Architecture

(2) Bicycles (3) Camping (4) Canoes (5) Cold showers

(6) Doing crossword puzzles (7) Playing chess

(8) Politics

(9) Public speaking

(10) Receiving criticism (11) Rugby

(12) Soccer

(13) Statistics

(14) Japan

(15) Taxes

(16) Taxidermy

(32)

APPENDIX B.1: COMPLETE QUESTIONNAIRES:

Version 1 - Condition 1: No reciprocity / no stress

(33)
(34)
(35)
(36)
(37)
(38)

APPENDIX 2.4 Complete Questionnaire version 4 - Condition 4: Reciprocity / stress

(39)
(40)
(41)
(42)
(43)
(44)

APPENDIX 3.1 Residual analysis, results test for assumptions ANOVA

Residual analysis were performed to test for the assumptions of the two-way ANOVA analysis. Outliers were assessed by inspection of a boxplot, normality was assessed using Shapiro-Wilk's test, skewness and kurtosis values and inspection of the histograms and Q-Q plots for each cell of the design and homogeneity of variances was assessed by Levene's test.

Starting out with running a two-way ANOVA on the original data, thereby creating the new variables: ‘predicted values’ (the mean of each cell of the design), ‘residuals’ (the differences between the predicted value and the actual, observed value), and the studentized residuals (the differences between the predicted value and the actual, observed value divided by its standard error (Draper & Smith, 1998).

Outliers. Based on the residual variable values we found 6 outliers after inspection of the box plots per condition of the experiment as these contained values greater than 1.5 box- lengths from the edge of the box, see table 5 for results.

No stress condition Stress condition Total

No-reciprocity appeal

Condition 1 (no-reciprocity/no stress)

No outliers

Condition 2 (no-reciprocity/stress)

Four outliers:

P

articipant 1055, 1082, 1090 and 1093, all showing a residual value

of 1.93 indicating an original value of 30

minutes

n = 58

Reciprocity appeal

Condition 3 (reciprocity/no stress)

No outliers

Condition 4 (reciprocity/stress)

T

wo outliers:

Participant 1018, 1031 showing a residual value

of 4.21 indicating an original value 60 minutes

n = 58

Total n = 60 n = 56 n = 116

Table 5: Outliers per condition

Furthermore, by inspection of the studentized residuals four potential outliers were identified: participant 1063 and 1110 with studentized residuals of 3,15 and participant 1018 and 1031 with studentized residuals of 3.29, as these values are greater than 3 standard deviations.

Normality of distribution. The assumption of normality for ‘amount of time willing to help' scores was satisfied for none of the group combinations of stress and reciprocity, as assessed by Shapiro-Wilk's test (p < .05). Moreover, by inspection of the skewness and kurtosis, it seemed that the distributions of the ‘amount of time willing to help scores’ for conditions 3 and 4, were non-normal, as these proved to be (positively) skewed (z > 2.58).

Lastly, by visual inspection of the histograms and Q-Q plots, it seemed the ‘amount of time

willing to help’ scores were approximately normal distributed for each condition.

Referenties

GERELATEERDE DOCUMENTEN

Afterwards, more specific questions were asked as well like what they thought of the current content and layout, what data they select for the eOverdracht and how they get this data

Dans ce cadre, qui constitue le premier chapitre de ce mémoire, nous allons étudier le primitivisme, un « mouvement » théorique qui porte sur la représentation du

foundational level of the professional ethic of care. we have no doubt that, while these two threads are insufficient to conceive a complete political ethic of care, they

However, if people regard reasons for indulgence as more compelling when behavior is more tempting, licensing research should in our view also focus on the possibility that it

The International Covenant on Civil &amp; Political Rights and the International Covenant on Economic, Social &amp; Cultural Rights (to which the United Kingdom and Argentina are

Schoolholidayclub.co.uk searches all the major holiday suppliers, so they promise to offer as good value as visiting your local travel agent.. If the school you want to

Also, it provides knowledge about the occurrence of UPB which is important due to the detrimental effects UPB can cause (Askew et al., 2015). The results of this present research

Disposing of inventory namely results in salvage revenue and reduced holding cost (Willoughby, 2010). They investigated problems in which the item is kept in assortment. Thus,