Rule Following and Cooperation - A within-subjects design

2024-12-10

2024-12-11

The number of tokens placed in the blue bucket is the primary outcome of interest as it indicates the level of subjects’ rule following and cooperative behaviour, respectively. Within the CRISP paradigm this is our measure of Compliance/Cooperation. We elicit subjects’
conditional allocations to the blue bucket via the strategy method to capture their Conditional Preferences and follow the protocol of Thoni and Volk (2018) in classifying subjects’ propensity to cooperate (i.e., their attitudes in the ABC approach). We elicit subjects Normative and Descriptive Beliefs (jointly their Social Expectations) and their personal norms (capturing Respect for Rules).

Subjects will be randomly allocated to one treatment, which varies the presence and nature of a rule and the presence of a public good (treatments use an adapted version of the task introduced in Kimbrough and Vostroknutov (2018)). The five between-subjects treatments are
outlined below. For all subjects, the within-subjects element of the experiment is completed for their assigned treatment.

Common Elements: In all treatment’s subjects receive an endowment of 5 tokens which they can allocate between a yellow and blue bucket. The yellow bucket always offers a higher return to self.

1. No Rule: In this treatment there is no rule or no public good component. We use this treatment as a benchmark for comparison, allowing us to determine the relative importance of factors within the CRISP framework for explaining costly rule following and voluntary cooperation by adding elements to this treatment.
2. Baseline RFT: This treatment introduces a nonenforced personally costly rule stating, “The rule is to place the tokens in the blue bucket”, allowing us to determine the role respect for rules plays in driving behaviour.
3. Baseline PGG: This treatment introduces a linear public goods game structure to the task as subjects are (post-hoc) placed in groups of 4. Incentives (i.e., return to self from own action) are equalised in this social task with respect to the individualistic tasks. The marginal per capita return is set at 0.5 with the allocations of all group members to the blue bucket doubled and distributed equally among the four group
members irrespective of their own contribution.
4. PGG Rule: This treatment adds a nonenforced rule to the Baseline PGG treatment. The wording of the rule is the same as in the Baseline RFT treatment, stating “The rule is to place the tokens in the blue bucket”.
5. PGG Social Rule: This treatment modifies the wording of the rule in PGG Rule to include an explicitly stated purpose, stating “The rule is to place the tokens in the blue bucket because it helps the other group members”.

For each of the above between-subjects treatments, the experiment uses a within–subjects design, applying the ABC method (Fischbacher and Gaechter, 2010; Fischbacher, Gaechter and Quercia, 2012; Gaechter, Kolle and Quercia, 2017; Isler et al., 2021; Gaechter, Kolle and Quercia, 2022; Weber et al., 2023; Gaechter and Marino Fages, 2023) or analogue of it to uncover the role of normative and descriptive beliefs and attitudes in fostering cooperation. Modifying the protocol of Gaechter, Kolle and Quercia (2017) the experiment comprises of four components.
The order in which subjects see the first two components will be randomised.

1. Behaviour Strategy Method Experiment (elicitation of behavioural attitudes)
In this first component of the experiment subjects make one unconditional choice, and fill out one contribution table in which they can condition their actions on the behaviour of others (elicited via the strategy method as in Fischbacher, Gaechter and Fehr (2001)). The structure of this component of the experiment is outlined below.
Timeline of Behaviour Strategy Method Experiment:
a. Unconditional Choice: Subjects choose how to allocate their five tokens between the blue and yellow bucket.
b. Conditional Choice – Behaviour: Subjects fill out a contribution schedule, conditioning their allocation to the blue bucket on the behaviour of 3 other randomly selected Prolific subjects.
Once we have collected all responses, subjects will be matched post-hoc to form groups of four.

2. Belief Strategy Method Experiment (elicitation of normative attitudes)
In this second component of the experiment subjects make one unconditional choice, and fill out one contribution table in which they can condition their actions on the normative beliefs (i.e., personal norm - see below) of others (elicited via the strategy method as in Fischbacher, Gaechter and Fehr (2001)). The structure of this component of the experiment is as follows. After making their unconditional choice, subjects provide their unincentivised personal norm by indicating the most appropriate allocation to the blue bucket.
Timeline of Belief Strategy Method Experiment:
a. Unconditional Choice: Subjects choose how to allocate their five tokens between the blue and yellow bucket.
b. Elicitation of Personal Norm: Subjects indicate which allocation to the blue bucket they think is most appropriate.
c. Conditional Choice – Normative Beliefs: Subjects fill out a contribution schedule, conditioning their allocation to the blue bucket on the personal normative beliefs of 3 other randomly selected Prolific subjects.
Once we have collected all responses, subjects will be matched post-hoc to form groups of four. Incentivisation of the above strategy method components of the experiment will be analogous to the one-shot study of Fischbacher, Gaechter and Quercia (2012) and is explained below.

3. Direct Response Experiment (elicitation of beliefs and effective contribution)
The third component of the experiment comprises a one–shot task which allows us to elicit subjects’ beliefs and effective contributions. In this task subjects make a contribution choice (their effective contribution - note that we use the term ‘contribution’ to refer to the number of tokens subjects place in the blue bucket (i.e., the action that represents a public good contribution in the PGG treatments or payoff-analogous to that in
the No Rule and Baseline RFT treatments)). We ask subjects to briefly explain their chosen allocation to the blue bucket in the direct response experiment. We then elicit their personal norm about the most appropriate allocation to the blue bucket, before eliciting their beliefs about the behaviour and personal norms (i.e., their descriptive and normative beliefs) of the other three randomly selected Prolific subjects. As in the previous components, once we have collected all responses, subjects will be matched post-hoc to form groups of four.
Thus, for all subjects we also have two forms of beliefs – descriptive beliefs about the behaviour of others and normative beliefs about the personal norms of others, which we can use separately and in conjunction when applying the ABC method to analyse the relative importance of normative and descriptive attitudes and beliefs in facilitating costly rule following and cooperation.

4. Elicitation of Personal Normative Beliefs, Social Normative Beliefs, and Descriptive Beliefs
As an additional component we follow the protocol of Basic and Verrina (2024) in eliciting subjects’ personal normative beliefs and social normative beliefs about all possible allocations to the blue bucket. Basic and Verrina (2024) consider personal norms as individually held standards of appropriate behaviour, while social norms refer to societal standards of appropriate behaviour. In addition, we also elicit subjects’ descriptive beliefs using a method akin to Eckel et al. (2021) and Heinicke, K¨onig-Kersting and Schmidt (2022).

5. Procedural Considerations
Subjects will be advised that the structure of the experiment consists of three parts, consisting of five tasks in total. Tasks 1 and 2 form Part 1 of the experiment, Tasks 3 and 4 form Part 2 of the experiment, while Task 5 constitutes Part 3:
Task 1. their unconditional choice in the first strategy method component
Task 2. their conditional choices in the first strategy method component
Task 3. their unconditional choice in the second strategy method component
Task 4. their conditional choices in the second strategy method component
Task 5. the direct response component

We also elicit personal normative beliefs, social normative beliefs, and descriptive beliefs as follow up questions, after subjects have completed Task 5. To ensure incentive compatibility subjects’ will be paid for one randomly chosen part they complete in the experiment. Subjects will be randomly rematched post-hoc for each part of the experiment. Within Part 1, for one randomly chosen Prolific participant their conditional choice in Task 2 will be payoff relevant. For the other 3 Prolific subjects, their unconditional choice in Task 1 is payoff relevant. If Part 2 is selected, bonus payments
are determined analogously. If Part 3 is chosen to determine subjects’ bonus they will be paid based on their choice in the direct response experiment and the accuracy of one of their guesses about the behaviour and beliefs of others in Task 5. All subjects will be paid for the accuracy of either one of their social normative beliefs or one of their descriptive beliefs in the additional questions we pose about Task 5.

Subjects will be randomly allocated into treatments by the LIONESS Lab software (Giamattei et al., 2020). We randomise subjects until each treatment has reached the desired sample size of 140. Once a treatment has achieved a sample size of 140 future subjects will no longer be randomised to that treatment.

Subjects are randomly allocated to one treatment. The order in which they complete the Behaviour Strategy Method Experiment and the Belief Strategy Method Experiment is also randomised.

Yes

As each subject makes multiple decisions, we will cluster at the level of the individual.

We will collect 700 observations in total, 140 per treatment (i.e., 140 in No Rule, 140 in Baseline RFT, 140 in Baseline PGG, 140 in PGG Rule, and 140 in PGG Social Rule). Exclusion Criteria: We will exclude any incomplete data, and consider the first 140 complete submissions in each treatment from subjects who submitted a completion code on Prolific. Subjects’ who answer comprehension questions incorrectly 2 or more times in No Rule and Baseline RFT (3 or more times in Baseline PGG, PGG Rule, and PGG Social Rule) are screened out of the study. In our analysis we will exclude: (i) subjects’ who self-report that hey did not pay attention during the study* (ii) subjects in Baseline PGG, PGG Rule, or PGG Social Rule who completed the study in 10 minutes or less (iii) subjects in No Rule or Baseline RFT who completed the study in 8 minutes or less. After applying these exclusion criteria if the sample size in any treatment drops below 140 we will recruit additional sub- jects to make up the difference in the impacted treatment (e.g., if the sample drops to 139 in No Rule, we will recruit 1 additional subject fulfilling the aforementioned non-exclusion criteria to the No Rule treatment). *This is based on subjects’ response to the following question: “The following question does not affect your payment, and is only used for data quality purposes. Please indicate the extent to which you agree with the statement: Throughout the study I was paying sufficient attention.” with response options Strongly Disagree, Disagree, Agree, Strongly Agree. We will exclude those who do not indicate Agree or Strongly Agree.

We will collect 700 observations in total, 140 per treatment (i.e., 140 in No Rule, 140 in Baseline RFT, 140 in Baseline PGG, 140 in PGG Rule, and 140 in PGG Social Rule).