ask: How important is it to ask pointed questions?

In JECheadle/RSOC317L: Data Sets and Supplemental Functions from David Diez's 'OpenIntro' Textbook

Description

Something is wrong with this data set.

In this experiment, each individual was asked to be a seller of an iPod (a product commonly used to store music on before smart phones...). They participant received $10 + 5% of the sale price for participating. The iPod they were selling had frozen twice in the past inexplicably but otherwise worked fine. The prospective buyer starts off and then asks one of three final questions, depending on the seller's treatment group.

The three possible questions:

• General: What can you tell me about it?

• Positive Assumption: It doesn't have any problems, does it?

• Negative Assumption: What problems does it have?

The outcome variable is whether or not the participant discloses or hides the problem with the iPod.

Usage

data("ask")

Format

A data frame with 219 observations on the following 3 variables.

question.class

The type of question: general, pos_assumption, and neg_assumption.

question

The question corresponding to the question.class

response

The classified response from the seller, either disclose or hide.

Source

Minson JA, Ruedy NE, Schweitzer ME. There is such a thing as a stupid question: Question disclosure in strategic communication. http://opim.wharton.upenn.edu/DPlab/papers/workingPapers/Minson_working_Ask%20(the%20Right%20Way)%20and%20You%20Shall%20Receive.pdf

Examples

data(ask)
(ask.table <- table(ask[, c("question.class", "response")]))
(test <- chisq.test(ask.table))
barplot(t(ask.table), col = COL[1:2])

# Check the test's assumption around sufficient expected observations
# per table cell.
names(test)
test$expected

JECheadle/RSOC317L documentation built on May 15, 2019, 4:02 a.m.