scholarly journals Estimating and Using Individual Marginal Component Effects from Conjoint Experiments

2021 ◽  
pp. 1-14
Author(s):  
Kirill Zhirkov
Keyword(s):  
Conjoint Analysis ◽  
Task Design ◽  
Individual Level ◽  
Individual Preferences ◽  
Political Preferences ◽  
Conjoint Experiment ◽  
The Individual ◽  
Heterogeneity Of Preferences ◽  
Additional Assumptions

Abstract Conjoint experiments are quickly gaining popularity as a vehicle for studying multidimensional political preferences. A common way to explore heterogeneity of preferences estimated with conjoint experiments is by estimating average marginal component effects across subgroups. However, this method does not give the researcher the full access to the variation of preferences in the studied populations, as that would require estimating effects on the individual level. Currently, there is no accepted technique to obtain estimates of individual-level preferences from conjoint experiments. The present paper addresses this gap by proposing a procedure to estimate individual preferences as respondent-specific marginal component effects. The proposed strategy does not require any additional assumptions compared to the standard conjoint analysis, although some changes to the task design are recommended. Methods to account for uncertainty in resulting estimates are also discussed. Using the proposed procedure, I partially replicate a conjoint experiment on immigrant admission with recommended design adjustments. Then, I demonstrate how individual marginal component effects can be used to explore distributions of preferences, intercorrelations between different preference dimensions, and relationships of preferences to other variables of interest.

Levels of Aggregation in Conjoint Analysis: An Empirical Comparison

1980 ◽  
Vol 17 (4) ◽  
pp. 516-523 ◽  
Author(s):  
William L. Moore
Keyword(s):  
Conjoint Analysis ◽  
Predictive Power ◽  
Aggregate Level ◽  
Segmentation Method ◽  
Empirical Comparison ◽  
Individual Level ◽  
Aggregate Analysis ◽  
The Individual ◽  
Level Analysis

Two segmented methods of performing conjoint anal/sis, clustered and componential segmentation, are compared with each other as well as with individual level and totally aggregate level analyses. The two segmented methods provide insights to the data that (1) are not obtainable at the aggregate level and (2) are in a form that is more easily communicated than the information from the individual level analysis. The predictive power of the clustered segmentation method is higher than that of componential segmentation, and both are superior to the aggregate analysis but inferior to individual level analysis.

Digging Deeper into the Gender Gap: Gender Salience as a Moderating Factor in Political Attitudes

2017 ◽  
Vol 50 (2) ◽  
pp. 559-578 ◽  
Author(s):  
Amanda Bittner ◽  
Elizabeth Goodyear-Grant
Keyword(s):  
Gender Identity ◽  
Gender Gap ◽  
Political Attitudes ◽  
Original Data ◽  
Political Behaviour ◽  
Individual Level ◽  
Political Preferences ◽  
The Individual ◽  
Election Studies ◽  
The Impact

AbstractWe know how sex (rather than gender) structures political preferences, but researchers rarely take into account the salience or importance of gender identity at the individual level. The only similar variable for which salience is commonly taken seriously is partisanship, for which direction and importance or strength are both considered imperative for measurement and analysis. While some scholars have begun to look at factors that may influence intragroup differences, such as feminism (Conover, 1988), most existing research implicitly assumes gender salience is homogenous in the population. We argue that both the content of gender identity (that is, what specifically is gender identity, as opposed to sex) as well its salience should be incorporated into analyses of how gender structures political behaviour. For some, gender simply does not motivate behaviour, and the fact that salience moderates the impact of gender on behaviour requires researchers to model accordingly. Using original data from six provincial election studies, we examine a measure of gender identity salience and find that it clarifies our understanding of gender's impact on political attitudes.

Structural Reliability and Stability of Nonmetric Conjoint Analysis

1982 ◽  
Vol 19 (2) ◽  
pp. 199-207 ◽  
Author(s):  
Naresh K. Malhotra
Keyword(s):  
Conjoint Analysis ◽  
Structural Reliability ◽  
Standard Errors ◽  
Relative Importance ◽  
Substantial Variation ◽  
Individual Level ◽  
Structural Perturbation ◽  
Estimated Parameters ◽  
Treatment Conditions ◽  
The Individual

Structural reliability and stability of nonmetric conjoint analysis are examined under conditions of severe structural perturbation and substantial variation in the number of stimulus profiles. The individual-level part worth functions are jackknifed. The jackknifed parameters, derived relative importance weights, and standard errors of estimated parameters are examined across the different treatment conditions. The results indicate that conjoint analysis is a fairly robust procedure for assessing an individual's preferences.

Model Misspecification in Multiattribute Parameter Estimation

1981 ◽  
Vol 18 (1) ◽  
pp. 87-93 ◽  
Author(s):  
Frank J. Carmone ◽  
Paul E. Green
Keyword(s):  
Parameter Estimation ◽  
Experimental Design ◽  
Conjoint Analysis ◽  
Model Misspecification ◽  
Utility Functions ◽  
Individual Level ◽  
Data Points ◽  
The Individual ◽  
Main Effects

Most applications of conjoint analysis have emphasized main-effects models, largely because fewer data points are needed to fit that type of model at the individual level. The authors suggest that such simplifications can lead to poor predictions when the underlying utility functions depart from the simplicity of a main-effects model. They also show how compromise designs, which allow orthogonal estimation of selected two-way interactions (as well as main effects), can provide a more general experimental design in cases where a specified set of two-way interactions is suspected.

scholarly journals Measuring the Impact of Interdependence on Individuals During Collaborative Problem-Solving

2021 ◽  
Vol 8 (1) ◽  
pp. 75-94
Author(s):  
Zachari Swiecki
Keyword(s):  
Problem Solving ◽  
Task Design ◽  
Collaborative Problem Solving ◽  
Individual Level ◽  
Collaborative Problem ◽  
Novel Method ◽  
The Individual ◽  
The Impact ◽  
Level Of Analysis ◽  
Over Time

Collaboration analytics often focuses on assessing and monitoring individuals during collaborative problem-solving (CPS). A defining feature of CPS is the interdependence that exists between individuals when they work together — that is, how they respond to and influence one another over time. While models that account for the impact of interdependence at the individual level of analysis (interdependent models) exist, they are often highly complex. This complexity makes them potentially difficult to use in assessments and systems that need to be explainable for educators, learners, and other researchers. Measures of the impact of interdependence at the individual level of analysis could inform decisions as to whether interdependent models should be used, or whether simpler models will suffice. Such measures could also be used to investigate specific questions about interdependence in collaborative settings. In this paper, I present a novel method of measuring the impact of interdependence on individuals using epistemic network analysis. To provide evidence of the validity of the measure, I compare it to qualitative findings that describe the impact of interdependence on individuals participating in team training scenarios. To demonstrate the value of the measure, I use it to assess the impact of interdependence in these data overall and to test hypotheses regarding the collaborative task design. My results suggest that the measure can distinguish between individuals who have been impacted by interdependence differently, that interdependence is impactful in these data overall, and that aspects of the task design may have affected how some individuals were impacted by interdependence.

The Political Mobilization of Ethnic and Religious Identities in Africa

2014 ◽  
Vol 108 (4) ◽  
pp. 801-816 ◽  
Author(s):  
JOHN F. McCAULEY
Keyword(s):  
Political Mobilization ◽  
The Political ◽  
Linguistic Context ◽  
World Religions ◽  
Individual Level ◽  
Political Preferences ◽  
Group Members ◽  
Level I ◽  
Using Data ◽  
The Individual

When elites mobilize supporters according to different cleavages, or when individuals realign themselves along new identity lines, do their political preferences change? Scholars have focused predominantly on the size of potential coalitions that leaders construct, to the exclusion of other changes that might occur when one or another identity type is made salient. In this article, I argue that changes in the salience of ethnicity and religion in Africa are associated with variation in policy preferences at the individual level. I test this claim empirically using data from a framing experiment in Côte d’Ivoire and Ghana. By randomly assigning participants to either a religious or an ethno-linguistic context, I show that group members primed to ethnicity prioritize club goods, the access to which is a function of where they live. Otherwise identical individuals primed to religion prioritize behavioral policies and moral probity. These findings are explained by the geographic boundedness of ethnic groups and the geographic expansiveness of (world) religions in the study area.

Differential Effects of Competence and Morality on Self-Esteem at the Individual and the Collective Level

2020 ◽  
Vol 51 (3) ◽  
pp. 183-198
Author(s):  
Wiktor Soral ◽  
Mirosław Kofta
Keyword(s):  
Self Esteem ◽  
Differential Effects ◽  
Individual Level ◽  
The Subject ◽  
The Individual

Abstract. The importance of various trait dimensions explaining positive global self-esteem has been the subject of numerous studies. While some have provided support for the importance of agency, others have highlighted the importance of communion. This discrepancy can be explained, if one takes into account that people define and value their self both in individual and in collective terms. Two studies ( N = 367 and N = 263) examined the extent to which competence (an aspect of agency), morality, and sociability (the aspects of communion) promote high self-esteem at the individual and the collective level. In both studies, competence was the strongest predictor of self-esteem at the individual level, whereas morality was the strongest predictor of self-esteem at the collective level.

Metonymy in Dance: Ballet Bunheads Take a Cognitive Turn

2019 ◽  
Vol 37 (1) ◽  
pp. 18-34
Author(s):  
Edward C. Warburton
Keyword(s):  
Identity Formation ◽  
Cognitive Science ◽  
Artistic Development ◽  
Individual Level ◽  
Figures Of Speech ◽  
Self Image ◽  
Level I ◽  
Dance Educators ◽  
The Individual ◽  
Metaphor And Metonymy

This essay considers metonymy in dance from the perspective of cognitive science. My goal is to unpack the roles of metaphor and metonymy in dance thought and action: how do they arise, how are they understood, how are they to be explained, and in what ways do they determine a person's doing of dance? The premise of this essay is that language matters at the cultural level and can be determinative at the individual level. I contend that some figures of speech, especially metonymic labels like ‘bunhead’, can not only discourage but dehumanize young dancers, treating them not as subjects who dance but as objects to be danced. The use of metonymy to sort young dancers may undermine the development of healthy self-image, impede strong identity formation, and retard creative-artistic development. The paper concludes with a discussion of the influence of metonymy in dance and implications for dance educators.

Sign in / Sign up

Export Citation Format

Share Document