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.

Voting Outcomes in the 1984 Democratic Party Primaries and Caucuses

1987 ◽  
Vol 81 (1) ◽  
pp. 67-84 ◽  
Author(s):  
T. Wayne Parent ◽  
Calvin C. Jillson ◽  
Ronald E. Weber
Keyword(s):  
Democratic Party ◽  
Aggregate Level ◽  
Individual Level ◽  
Primary Election ◽  
Electoral Outcomes ◽  
The Individual ◽  
Scholarly Inquiry ◽  
Level Analysis ◽  
Level Of Analysis

Scholarly inquiry concerning influences on electoral outcomes in the presidential nomination process, though extensive, has been conducted almost exclusively with data collected at the individual level of analysis. The Michigan model of normal vote analysis suggests that long-term influences measured at the aggregate level, such as the sociodemographic, economic, and ideological characteristics of the states, are also important in determining electoral outcomes. We present an aggregate-level analysis of state characteristics that affected the Hart, Jackson, and Mondale vote proportions in the 1984 Democratic caucuses and primaries. Our primary election models explain between 65% and 83% of the variance in candidate vote shares, with sociodemographic and economic factors as the leading indicators. In the caucuses, we find that campaign spending and sociodemographic influences are dominant in models that explain between 38% and 81% of the variance. We conclude with a brief discussion of what our findings mean for future Democratic candidates.

A Critical Review of Theories and Models of Technology Adoption and Acceptance in Information System Research

2015 ◽  
Vol 6 (4) ◽  
pp. 58-77 ◽  
Author(s):  
Ali Tarhini ◽  
Nalin Asanka Gamagedara Arachchilage ◽  
Ra'ed Masa'deh ◽  
Muhammad Sharif Abbasi
Keyword(s):  
Information System ◽  
Technology Adoption ◽  
Environmental Factors ◽  
Critical Review ◽  
Predictive Power ◽  
Theoretical Models ◽  
System Research ◽  
Information System Research ◽  
Individual Level ◽  
The Individual

Previous research shows that selecting an appropriate theory or model has always remained a critical task for IS researchers. To the best of the authors' knowledge, there are few papers that review and compare the acceptance theories and models at the individual level. Hence, this article aims to overcome this problem by providing a critical review of eight of the most influential theories that have been used to predict and explain human behaviour towards adoption of various technologies at the individual level. This article also summarizes their evolution; highlight the key constructs, extensions, strengths, and criticisms from a selective list of published articles appeared in the literature related to IS. This review provides a holistic picture for future researchers in selecting appropriate single/multiple theoretical models/constructs based on their strengths and weaknesses and in terms of predictive power and path significance. It is concluded that a well-established theory should consider the personal, social, cultural, technological, organizational and environmental factors

scholarly journals Frankly, We Do Give a Damn

2017 ◽  
Vol 8 (7) ◽  
pp. 816-826 ◽  
Author(s):  
Gilad Feldman ◽  
Huiwen Lian ◽  
Michal Kosinski ◽  
David Stillwell
Keyword(s):  
Social Interactions ◽  
Positive Relationship ◽  
Real Life ◽  
Linguistic Analysis ◽  
The Other ◽  
Aggregate Level ◽  
Individual Level ◽  
Common Causes ◽  
The Individual ◽  
The Relationship

There are two conflicting perspectives regarding the relationship between profanity and dishonesty. These two forms of norm-violating behavior share common causes and are often considered to be positively related. On the other hand, however, profanity is often used to express one’s genuine feelings and could therefore be negatively related to dishonesty. In three studies, we explored the relationship between profanity and honesty. We examined profanity and honesty first with profanity behavior and lying on a scale in the lab (Study 1; N = 276), then with a linguistic analysis of real-life social interactions on Facebook (Study 2; N = 73,789), and finally with profanity and integrity indexes for the aggregate level of U.S. states (Study 3; N = 50 states). We found a consistent positive relationship between profanity and honesty; profanity was associated with less lying and deception at the individual level and with higher integrity at the society level.

Individual Components of Three Inequality Measures for Analyzing Shapes of Inequality

2019 ◽  
pp. 004912411987596
Author(s):  
Tim Futing Liao
Keyword(s):  
Sociological Research ◽  
Aggregate Level ◽  
Current Population Survey Data ◽  
Single Measure ◽  
Generalized Entropy ◽  
Individual Level ◽  
Inequality Measures ◽  
Entropy Measures ◽  
Individual Contributions ◽  
The Individual

In common sociological research, income inequality is measured only at the aggregate level. The main purpose of this article is to demonstrate that there is more than meets the eye when inequality is indicated by a single measure. In this article, I introduce an alternative method that evaluates individuals’ contributions to inequality as well as the between-group and within-group components of these individual contributions. I first highlight three common inequality measures, the Gini index and two generalized entropy measures—Theil’s T and Theil’s L indices—by presenting their individual components as a method for evaluating inequality. Five artificial data examples illustrate the use of these individual components first. An empirical analysis of the 2007 and 2017 Current Population Survey data then focuses on the differences in inequality revealed by such individual inequality components between the 2007 and 2017. The individual-level inequality measures can reveal patterns of inequality concealed by single measures at the aggregate level. In particular, the Gini individual measures differentiate cases better than the generalized entropy measures and tend to have smaller standard errors in a regression analysis.

Trustees, Delegates, and Responsiveness in Comparative Perspective

2016 ◽  
Vol 50 (6) ◽  
pp. 766-793 ◽  
Author(s):  
Shaun Bowler
Keyword(s):  
Academic Research ◽  
Large Body ◽  
European Social Survey ◽  
Aggregate Level ◽  
Social Survey ◽  
Individual Level ◽  
Conceptual Distinction ◽  
Citizen Preferences ◽  
The Individual ◽  
Government Responsiveness

A large body of aggregate-level work shows that government policies do indeed respond to citizen preferences. But whether citizens recognize that government is responsive is another question entirely. Indeed, a prior question is whether or not citizens value responsiveness in the way that academic research assumes they should in the first place. Using comparative data from the European Social Survey, this article examines how citizens see government responsiveness. We show that several key assumptions of the aggregate-level literature are met at the individual level. But we also present results that show that attitudes toward representation and responsiveness are colored, sometimes in quite surprising ways, by winner–loser effects. In a finding that stands in some contrast to the normative literature on the topic, we show that these sorts of short-term attitudes help shape preferences for models of representation. In particular, we show that the distinction between delegates and trustees is a conceptual distinction that has limits in helping us to understand citizen preferences for representation.

scholarly journals The disproportion of crowd wisdom: The impact of status seeking on Yelp reviews

PLoS ONE ◽  
2021 ◽  
Vol 16 (6) ◽  
pp. e0252157
Author(s):  
Chao Yu ◽  
Drew Margolin
Keyword(s):  
Wisdom Of Crowds ◽  
Aggregate Level ◽  
Variety Seeking ◽  
Individual Level ◽  
Price Levels ◽  
Crowd Wisdom ◽  
Status Seeking ◽  
Restaurant Reviews ◽  
The Impact ◽  
Level Analysis

This study shows that while status seeking motivates people to participate in crowdsourcing platforms, it also negatively impacts the bedrock of crowdsourcing–wisdom of crowds. Using Yelp restaurant reviews in 6 cities, we found that motivations of status seeking lead people to review a greater variety of restaurants, and achieving status further encourages this variety seeking as well as the targeting of more expensive restaurants for review. The impact of this individual-level tendency is confirmed by our aggregate-level analysis which shows that restaurants with higher price levels, higher uniqueness levels, and a larger percentage of elite reviews tend to obtain enough reviews to generate wisdom of crowds sooner than other restaurants. This leads to a different kind of distortion to crowd wisdom: an over-representation of status-conferring products and an under-representation of products that are not status-worthy. The findings suggest the importance of studying sources of distortion that are endemic to crowdsourcing itself.

scholarly journals Can education be personalised using pupils’ genetic data?

2019 ◽  
Author(s):  
Tim T Morris ◽  
Neil M Davies ◽  
George Davey Smith
Keyword(s):  
Predictive Power ◽  
Predictive Accuracy ◽  
Educational Performance ◽  
Phenotypic Data ◽  
Individual Level ◽  
Polygenic Scores ◽  
Prior Achievement ◽  
Using Data ◽  
The Individual ◽  
Potential Tool

AbstractThe increasing predictive power of polygenic scores for education has led to their promotion by some as a potential tool for genetically informed policy. How well polygenic scores predict educational performance conditional on other phenotypic data is however not well understood. Using data from a UK cohort study, we investigated how well polygenic scores for education predicted pupils’ realised achievement over and above phenotypic data that are available to schools. Across our sample, prediction of educational outcomes from polygenic scores were inferior to those from parental socioeconomic factors. There was high overlap between the polygenic score and achievement distributions, leading to weak predictive accuracy at the individual level. Furthermore, conditional on prior achievement polygenic scores were not predictive of later achievement. Our results suggest that while polygenic scores can be informative for identifying group level differences, they currently have limited use for predicting individual educational performance or for personalised education.

scholarly journals Nonergodicity in Load and Recovery: Group Results Do Not Generalize to Individuals

2021 ◽  
pp. 1-9
Author(s):  
Niklas D. Neumann ◽  
Nico W. Van Yperen ◽  
Jur J. Brauers ◽  
Wouter Frencken ◽  
Michel S. Brink ◽  
...  
Keyword(s):  
Perceived Exertion ◽  
Rating Of Perceived Exertion ◽  
Group Level ◽  
Total Quality ◽  
Individual Level ◽  
Quality Of Recovery ◽  
Level Statistics ◽  
The Individual ◽  
Level Analysis

Purpose: The study of load and recovery gained significant interest in the last decades, given its important value in decreasing the likelihood of injuries and improving performance. So far, findings are typically reported on the group level, whereas practitioners are most often interested in applications at the individual level. Hence, the aim of the present research is to examine to what extent group-level statistics can be generalized to individual athletes, which is referred to as the “ergodicity issue.” Nonergodicity may have serious consequences for the way we should analyze, and work with, load and recovery measures in the sports field. Methods: The authors collected load, that is, rating of perceived exertion × training duration, and total quality of recovery data among youth male players of a professional football club. This data were collected daily across 2 seasons and analyzed on both the group and the individual level. Results: Group- and individual-level analysis resulted in different statistical outcomes, particularly with regard to load. Specifically, SDs within individuals were up to 7.63 times larger than SDs between individuals. In addition, at either level, the authors observed different correlations between load and recovery. Conclusions: The results suggest that the process of load and recovery in athletes is nonergodic, which has important implications for the sports field. Recommendations for training programs of individual athletes may be suboptimal, or even erroneous, when guided by group-level outcomes. The utilization of individual-level analysis is key to ensure the optimal balance of individual load and recovery.

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.

scholarly journals Predicting Violence in Merseyside: a Network-Based Approach Using No Demographic Information

2020 ◽  
Vol 4 (3-4) ◽  
pp. 89-102
Author(s):  
Paolo Campana ◽  
Andrea Giovannetti
Keyword(s):  
Regression Models ◽  
Predictive Power ◽  
Demographic Information ◽  
Data Set ◽  
Logistic Regression Models ◽  
Individual Level ◽  
The Individual ◽  
High Level ◽  
Very High ◽  
Individual Information

Abstract Purpose We explore how we can best predict violent attacks with injury using a limited set of information on (a) previous violence, (b) previous knife and weapon carrying, and (c) violence-related behaviour of known associates, without analysing any demographic characteristics. Data Our initial data set consists of 63,022 individuals involved in 375,599 events that police recorded in Merseyside (UK) from 1 January 2015 to 18 October 2018. Methods We split our data into two periods: T1 (initial 2 years) and T2 (the remaining period). We predict “violence with injury” at time T2 as defined by Merseyside Police using the following individual-level predictors at time T1: violence with injury; involvement in a knife incident and involvement in a weapon incident. Furthermore, we relied on social network analysis to reconstruct the network of associates at time T1 (co-offending network) for those individuals who have committed violence at T2, and built three additional network-based predictors (associates’ violence; associates’ knife incident; associates’ weapon incident). Finally, we tackled the issue of predicting violence (a) through a series of robust logistic regression models using a bootstrapping method and (b) through a specificity/sensitivity analysis. Findings We found that 7720 individuals committed violence with injury at T2. Of those, 2004 were also present at T1 (27.7%) and co-offended with a total of 7202 individuals. Regression models suggest that previous violence at time T1 is the strongest predictor of future violence (with an increase in odds never smaller than 123%), knife incidents and weapon incidents at the individual level have some predictive power (but only when no information on previous violence is considered), and the behaviour of one’s associates matters. Prior association with a violent individual and prior association with a knife-flagged individual were the two strongest network predictors, with a slightly stronger effect for knife flags. The best performing regressors are (a) individual past violence (36% of future violence cases correctly identified); (b) associates’ past violence (25%); and (c) associates’ knife involvement (14%). All regressors are characterised by a very high level of specificity in predicting who will not commit violence (80% or more). Conclusions Network-based indicators add to the explanation of future violence, especially prior association with a knife-flagged individual and association with a violent individual. Information about the knife involvement of associates appears to be more informative than a subject’s own prior knife involvement.

Sign in / Sign up

Export Citation Format

Share Document