Why Are There so Few Formal Measuring Instruments in Social and Political Research?

Heath, Anthony; Martin, Jean

doi:10.1002/9781118490013.ch3

Cited by 24 publications

(17 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It only means that the invariance of the items that we have examined is quite limited. Heath and Martin (1997) argue that there is a resource tradeoff between the ability of a survey project to invest in preliminary methodological work in scale construction and its ability to collect large representative samples. While that is a problem in any national survey, it is even a bigger problem in cross-national survey projects that need more resources and are required to handle more challenges, one of which is scale comparability.…”

Section: Resultsmentioning

confidence: 99%

Can we Rate Public Support for Democracy in a Comparable Way? Cross-National Equivalence of Democratic Attitudes in the World Value Survey

2010

View full text Add to dashboard Cite

In this study we examine the cross-cultural equivalence of two scales that measure attitudes toward democracy across 36 countries in the World Value Survey (WVS) 2000. We examine the equivalence of these scales in order to explore if we can meaningfully compare democratic attitudes across countries. Multiple group confirmatory factor analyses (MGCFA) is applied to answer this question. The analyses indicate that the scales may be compared but only to a certain extent and not across all the countries. We close this article by discussing the implications of the findings.

show abstract

Section: Resultsmentioning

confidence: 99%

Can we Rate Public Support for Democracy in a Comparable Way? Cross-National Equivalence of Democratic Attitudes in the World Value Survey

2010

View full text Add to dashboard Cite

show abstract

“…Nunnally (1967) argued that relatively low reliability coefficients (e.g., .50 or .60) are tolerable in the early stages of research, although he later adjusted this minimum level to .70 (Nunnally, 1978). Heath and Martin (1997) suggested that alpha values should be at least .60. In Study 1, two of the dimensions (idealism and individualism) were found to show alpha values at .60, which could be regarded as somewhat low but acceptable at this stage.…”

Section: Internal Consistency Of the Npis Subscalesmentioning

confidence: 99%

Measuring Military Identity: Scale Development and Psychometric Evaluations

Johansen¹,

Laberg²,

Martinussen³

2013

soc behav pers

View full text Add to dashboard Cite

Our purpose was to perform a psychometric evaluation of a new 33-item questionnaire developed in Norway. To evaluate it we assessed its internal consistency, performed an exploratory factor analysis, and investigated aspects of construct validity. We also examined test-retest reliability. A second purpose was to investigate whether or not individual level variables such as age, gender, or service were related to different military identities. In Study 1 we collected cross-sectional data from military personnel in the Norwegian Armed Forces (N = 317). In Study 2 we collected longitudinal data from students undertaking junior officer education (N = 238). We identified a 3-factor structure, comprising professionalism, individualism, and idealism. Internal consistency for the 3 subscales was acceptable ( = .60-.83). Test-retest reliability and construct validity were supported. We found professionalism to be significantly higher in the Army as compared to in the Navy and Air Force. We did not detect gender differences in terms of military identities, but we did detect small negative correlations between age and professionalism and between age and idealism.

show abstract

“…Heath and Martin (1997) caution against over-reliance on Cronbach's Alpha for the assessment of the reliability of "formal measuring instruments", suggesting that in some circumstances high Alpha values can be achieved where an underlying scale is not uni-dimensional. Therefore, in order to provide a more rigorous test of the stability of the structure and composition of the core value scales, Confirmatory Factor Models were developed, which allow for a formal test of the fit of the data with the proposed structure of social values.…”

Section: Reliability and The Core Value Scalesmentioning

confidence: 99%

A Reactive Core? The Configuration of Values in the British Electorate 1986–2007

Surridge

2012

Journal of Elections, Public Opinion & Parties

View full text Add to dashboard Cite

There has been a broad consensus since the early 1990s that the core values of the British electorate can be represented by two dimensions, a "left-right" dimension and an "authoritarian-liberal" dimension. There has been a keen interest in how the electorate's position on these scales has changed over time. However, there has been little attention paid to how the core values themselves may have changed in composition and effect over the two decades since the measures were first developed. This article uses data from the British Social Attitudes series from 1986 to 2007 and Confirmatory Factor Analyses to trace the structure of these core values over time and to evaluate if the way in which core values interact with key social and political attitudes has changed during this period.

show abstract

Why Are There so Few Formal Measuring Instruments in Social and Political Research?

Cited by 24 publications

References 20 publications

Can we Rate Public Support for Democracy in a Comparable Way? Cross-National Equivalence of Democratic Attitudes in the World Value Survey

Can we Rate Public Support for Democracy in a Comparable Way? Cross-National Equivalence of Democratic Attitudes in the World Value Survey

Measuring Military Identity: Scale Development and Psychometric Evaluations

A Reactive Core? The Configuration of Values in the British Electorate 1986–2007

Contact Info

Product

Resources

About