predictAttention: Predict Attention Among Respondents

In zieglerjef/openEnded: Tools for Implementing and Analyzing Open-Ended Responses as Manipulation Checks

Description

Estimate a linear regression in which attention is the outcome, and the user defines the characteristics used to predict attention.

Usage

predictAttention(
  dataframe = NULL,
  attention_formula = NULL,
  similarity_measures = c("jaccardSimilarity", "cosineSimilarity"),
  correct_vec = NULL,
  model_type = NULL,
  k = 3,
  up_down_weight = "down"
)

Arguments

dataframe	Dataframe from which you will select attention, predictors for attention, etc. for the regression model.
k	The penalty that you want to set for down-weighting inattentive respondents. Lower levels of k down-weight low attention participants more severely.
up_down_weight	Do you want to up-weight or down-weight? Default="down".
attention_predictors	Symbolic representation of the model to be estimated. This is written in "typical" R language (i.e. x1 + x2). Do not need to specify y or outcome variable, it's assigned internally.

Value

Object containing instrumental variable regression call, data, results, etc.

Author(s)

Jeffrey Ziegler (<jeffrey.ziegler[at]emory.edu>)

zieglerjef/openEnded documentation built on Nov. 30, 2020, 2:03 p.m.