'Replication of Study 3 of Wu, Moore, & Fitzsimons (2019)' (AsPredicted #24,999)
Author(s) Joseph Simmons (University of Pennsylvania) - [email protected] Leif Nelson (University of California, Berkeley) - [email protected]
Pre-registered on 2019/06/19 07:43 (PT)
1) Have any data been collected for this study already? No, no data have been collected for this study yet.
2) What's the main question being asked or hypothesis being tested in this study? This study is a replication of Study 3 in Wu, Moore, & Fitzsimons (2019), “Wine for the Table: Self-Construal, Group Size, and Choice for Self and Others,” published in the Journal of Consumer Research. This study examines the hypothesis laid out in the authors’ abstract: “interdependent consumers consistently make choices that balance self and others’ preferences, regardless of group size. In contrast, the choices of independent consumers differ depending on group size: for smaller groups, independents make choices that balance self and others’ preferences, while for larger groups, they make choices that more strongly reflect their own preferences.”
3) Describe the key dependent variable(s) specifying how they will be measured. There are a number of key measures.
Participants will imagine being in a book club and having to choose among a variety of booksellers’ offers. Each offer is for seven books, but they vary in terms of how many books are fiction vs. non-fiction. The first offer is for 7 fiction books and 0 non-fiction books, the second is for 6 fiction books and 1 non-fiction book, and so on, until the eighth and final offer is for 0 fiction books and 7 non-fiction books. Participants’ choice of offer is the key dependent variable (scored 1 = all fiction books, 8 = all non-fiction books).
Participants will report how accountable and responsible they would feel when choosing among the booksellers’ offers (on 7-point scales, ranging from 1 = not at all to 7 = very much). These two items will be averaged to form a composite measure of accountability. This measure will be used as a covariate in the main analysis.
Participants will indicate how often they read 14 book genres (1 = every day; 7 = once or twice a year), including the two critical categories of “non-fiction” and “general fiction.” These questions also include an option of “never” at the end of the scale, and participants’ endorsement of this option will be used as a basis for excluding them (as described in Question 6). For those who do not endorse the “never” option for either of these two items, we will use their response to these two questions as covariates in the main analysis.
Participants will indicate how much they like 14 book genres (1 = dislike very much; 7 = like very much), including the categories of “non-fiction” and “general fiction.” We will subtract participants’ liking of fiction from their liking of non-fiction to create a measure that captures “preference for non-fiction books.” This measure will be used as a predictor variable in the main analysis.
4) How many and which conditions will participants be assigned to? Participants will be randomly assigned to one cell of a 2 (prime) x 2 (group size) between-subjects design. The prime manipulation involves counting the number of pronouns in a paragraph that contains either exclusively plural pronouns or exclusively singular pronouns. The group size manipulation involves having participants imagine that they are in a book club with either 3 other people or 9 other people. We will use the same stimuli (and instructions) used by Wu et al. (2019).
For robustness, we are also introducing an additional “book set” manipulation that we expect to be inconsequential. In Wu et al.’s (2019) original study, which was conducted in the Fall of 2015, they showed participants four books, two fiction and two non-fiction, prior to making that choice. In our study, we will randomly assign participants to see either the same set of books that Wu et al. displayed or an updated set of books.
5) Specify exactly which analyses you will conduct to examine the main question/hypothesis. In our main analysis, we will regress participants’ book deal choice on (1) prime condition (-.5 = independent prime; +.5 = interdependent prime), (2) group size (-.5 = small; +.5 – large), (3) preference for non-fiction books (mean centered), (4)-(6) all possible 2-way interactions among those three variables, (7) the 3-way interaction between those variables, (8) the accountability measure (mean-centered), (9) how often they report reading general fiction (mean-centered), and (10) how often they report reading non-fiction (mean-centered). The key predictor is the 3-way interaction (7). If that interaction is significant, we will unpack it by running separate analyses within each prime condition.
We will also run this main analysis separately for the two book set conditions.
Although Wu et al. (2019) investigated other hypotheses in their Study 3, including mediation by pronoun usage in an essay that participants write, we will not do that here, unless for exploratory purposes.
6) Describe exactly how outliers will be defined and handled, and your precise rule(s) for excluding observations. We will exclude any participants who answer “never” to the question asking them how often they read “general fiction” or to the question asking them how often they read “non-fiction.” We will also exclude any participants who answer “I do not read this type of book” to the question asking them how much they like “general fiction” or to the question asking them how much they like “non-fiction.” Participants who do not answer all of these four measures will also be excluded.
7) How many observations will be collected or what will determine sample size? No need to justify decision, but be precise about exactly how the number will be determined. We will collect data from 800 U.S. MTurk participants. We will recruit MTurkers with a rating of 95 or above.
8) Anything else you would like to pre-register? (e.g., secondary analyses, variables collected for exploratory purposes, unusual analyses planned?) No
