Skip to main content Accessibility help

Comparative Research is Harder Than We Thought: Regional Differences in Experts’ Understanding of Electoral Integrity Questions

Published online by Cambridge University Press:  03 July 2019

Bruno Castanho Silva
Cologne Center for Comparative Politics, University of Cologne, Herbert-Lewin Str. 2, D-50931, Cologne, Germany. Email:
Levente Littvay
Department of Political Science, Central European University, Nador u. 9, 1051 Budapest, Hungary
E-mail address:


Expert evaluations about countries form the backbone of comparative political research. It is reasonable to assume that such respondents, no matter the region they specialize in, will have a comparable understanding of the phenomena tapped by expert surveys. This is necessary to get results that can be compared across countries, which is the fundamental goal of these measurement activities. We empirically test this assumption using measurement invariance techniques which have not been applied to expert surveys before. Used most often to test the cross-cultural validity and translation effects of public opinion scales, the measurement invariance tests evaluate the comparability of scale items across any groups. We apply them to the Perceptions of Electoral Integrity (PEI) dataset. Our findings suggest that cross-regional comparability fails for all eleven dimensions identified in PEI. Results indicate which items remain comparable, at least across most regions, and point to the need of more rigorous procedures to develop expert survey questions.

Copyright © The Author(s) 2019. Published by Cambridge University Press on behalf of the Society for Political Methodology. 

A large part of public opinion research is dedicated to studying whether individuals in different groups understand survey questions in similar ways (see Davidov, Muthén, and Schmidt Reference Davidov, Muthén and Schmidt2018, for an overview). However, while those using expert surveys have been concerned for a while with systematic differences in experts’ response patterns (e.g., Bakker et al. Reference Bakker, Edwards, Jolly, Polk, Rovny and Steenbergen2014), few studies to date applied measurement invariance (MI) techniques from public opinion research to expert surveys (for a recent exception, see Marquardt and Pemstein Reference Marquardt and Pemstein2018). Due to respondents’ homogeneity and usage of anchoring vignettes, one expects these problems to be minor or absent. Nevertheless, since measuring and comparing country averages is their main goal, if specialists on different countries or regions answer questions differently, comparison becomes impossible.

We show how MI techniques from mass surveys can be applied to analyze differing response patterns from experts using the Perceptions of Electoral Integrity dataset (PEI, Norris, Wynter, and Grömping Reference Norris, Wynter and Grömping2017). Results indicate that direct mean comparisons of countries’ electoral integrity across regions are biased across most integrity’s dimensions. We conclude with recommendations for those using this dataset and for the elaboration of better expert surveys in general.

1 Data

The PEI datasetFootnote 1 asks experts, since 2013, to evaluate whether national elections all around the world can be considered free and fair.Footnote 2 It has 49 questions capturing eleven dimensions of electoral integrity. With the exception of North Africa ( $n=41$ ), all regions have at least 100 expert respondents.Footnote 3 The dimensions of electoral integrity are: 1. Electoral Laws (3 questions); 2. Electoral Procedures (4); 3. Boundaries for Voting Districts (3); 4. Voter Registration (3); 5. Party and Candidate Registration (5); 6. Media Coverage (5); 7. Campaign Finance (5); 8. Voting Process (8); 9. Vote Count (5); 10. Voting Results (4); 11. Election Authorities (4). Responses have a 1–5 Likert scale, from “Strongly disagree” to “Strongly agree.” For example, on “Election Authorities,” the first indicator reads “The election authorities were impartial.”Footnote 4

2 Measurement Invariance

With Likert scales, MI is obtained if two respondents who have the same level of a latent construct give the same answer to a question measuring it, regardless of group membership (Millsap Reference Millsap2011).Footnote 5 We test two types of invariance: metric and scalar (Millsap Reference Millsap2011). Metric invariance means that an increase of 1 unit in the latent trait corresponds to the exact same increase in the response for any individual, regardless of group identity. It yields comparable coefficients when used in regression analysis. Scalar invariance requires, on top of metric, that two individuals with the same level of the construct would give the exact same answer. Scalar invariance is more demanding, but necessary for comparison of group means.

We test MI using multigroup confirmatory factor analysis (MGCFA, Jöreskog Reference Jöreskog1971). In a CFA model, indicators theorized to measure each dimension should load together into a single latent variable. MGCFA considers that observations might be nested into qualitatively different groups. It allows researchers to test if some estimated model parameters differ across the groups: for example, a factor loading may have one value for respondents from region A and another for B, while other model parameters are estimated as the same across all regions. If the measurement instrument is invariant across groups, freely estimated loadings should be similar: if the loading of an indicator is high for one region but low for another, it suggests that the items function differently. We apply MGCFA with three nested invariance models: configural, metric (factor loadings are constrained to be equal across groups) and scalar (loadings and intercepts are constrained to be equal across groups), testing invariance by comparing the $\unicode[STIX]{x1D712}^{2}$ -distributed difference in the -2*log-likelihoods.

The $\unicode[STIX]{x1D712}^{2}$ invariance test in MGCFA has been criticized for finding noninvariance even in the presence of substantively minor deviations (Oberski Reference Oberski2014). Therefore, we also present the Root Mean Squared Error of Approximation (RMSEA), and the Comparative Fit Index (CFI). These are less sensitive to large samples and indicate whether the more restricted models have acceptable fit. Next we use the Alignment method (Asparouhov and Muthén Reference Asparouhov and Muthén2014). This is an optimization approach to invariance which minimizes overall noninvariance of the model while allowing noninvariance of a few indicators or groups (Asparouhov and Muthén Reference Asparouhov and Muthén2014). While these may increase the odds of Type II errors (not finding noninvariance that exists), combining both methods gives us a good overview of actual (non)invariance levels.Footnote 6

We test MI in each of the eleven electoral integrity dimensions, each modeled as a latent variable with three to eight indicators as denoted in the codebook. Invariance is tested across two clustering options: five continents, or nine regions (defined by PEI itself). Instead of using the original responses, we run the analyses on DIF-corrected rank ordered responses obtained with PEI’s anchoring vignettes, following the nonparametric model by King et al. (Reference King, Murray, Salomon and Tandon2004).Footnote 7

Table 1. MGCFA test of measurement invariance.

$\unicode[STIX]{x1D6E5}\unicode[STIX]{x1D712}^{2}$ : difference between the configural and metric invariance models’ $\unicode[STIX]{x1D712}^{2}$ statistics; $df$ : difference in the number of free parameters between the two models, and $p$ -values for this difference. RMSEA and CFI from the metric invariance models; $\unicode[STIX]{x1D6E5}$ RMSEA and $\unicode[STIX]{x1D6E5}$ CFI are the difference in RMSEA and CFI from that model to the configural. All fit statistics are calculated with the Satorra and Bentler (Reference Satorra and Bentler2001) correction for robust estimation.

3 Results

Table 1 has results of metric invariance tests with MGCFA. We do not report scalar invariance because no dimension even approaches it.Footnote 8 For no dimensions of electoral integrity can we directly compare mean values of countries across regions. A few dimensions fare well in metric invariance: “Boundaries,” “Voter Registration,” and “Electoral authorities,” in both regions and continents. Most others not only display significant differences between the configural and metric models, but the models with constrained loadings have very poor fit, as denoted by RMSEAs greater than 0.05 and CFIs less than 0.95 (recommended thresholds for good fit by Hu and Bentler Reference Hu and Bentler1999). It is important to highlight that noninvariance remains after incorporating anchoring vignettes into the model,Footnote 9 confirming the finding by von Davier et al. (Reference von Davier, Shin, Khorramdel and Stankov2017) that vignette questions themselves can be understood differently by respondents in different groups, ultimately just moving the noninvariance problem one step further.

With the alignment analysis (Table 2), we find that noninvariance in some dimensions is restricted to a few items, and mostly concentrated in some regions. For example, the first indicator for “Electoral Laws” has a noninvariant intercept in Africa, while the third has noninvariant loading and intercept in Western Europe. These indicators, however, work well in all other continents or regions to measure the integrity of electoral laws in a cross-regionally comparable way. In these cases, users of PEI could drop problematic regions or items from their analyses. The best performing dimension is “Voter registration,” where all items have good cross-regional validity and can be used without problems in comparative research. Western Europe is the most problematic region overall, with the highest number of indicators that are noninvariant for loadings and intercepts. We may hypothesize that specialists in Western Europe have different anchors for inappropriate electoral conduct, resulting in different response styles even after accounting for DIF with anchoring vignettes.

Table 2. Alignment method: loadings and intercepts invariance.

Numbers indicate the standardized difference between the noninvariant group intercept and the weighed average intercept between invariant groups. Continents: 1: Africa; 2: Americas; 3: Asia; 4: Europe; 5: Oceania. Regions: 1: East & Southern Africa; 2: West & Central Africa; 3: East Asia & Pacific; 4: South Asia; 5: Eastern Europe; 6: Western Europe; 7: Middle East; 8: North Africa; 9: Americas.

Noninvariance of indicators’ intercepts is considered the least problematic, since if loadings are invariant one can still get unbiased regression estimates. To check the magnitude of the (smallest possible) problem we look at the size of noninvariance for intercepts. Numbers in the light gray cells are the standardized difference between that indicator’s intercept for that group and the weighed average intercept for invariant groups. Looking again at the noninvariant intercept for the third indicator in “Boundaries” in Europe, the estimate means that, for an election with the exact same level of integrity of boundaries for voting districts, an European respondent would give an answer 0.21 standard deviations below the average of respondents from other regions. There are larger differences in other dimensions and regions: for example, experts in Africa evaluating whether “fraudulent votes were cast” (“Voting Process”) would give answers 0.42 SDs below the average of other respondents evaluating exactly the same elections. Considering how these questions are aggregated, biases add up to substantively different evaluations of elections across regions.

4 Conclusion

In this letter we perform MI tests in the PEI dataset (Norris, Wynter, and Grömping Reference Norris, Wynter and Grömping2017), and show that experts have different response styles depending on the region where their countries of specialty are located. These results reinforce findings that one should not use aggregate means of expert surveys (Marquardt and Pemstein Reference Marquardt and Pemstein2018). The problem persists after using anchoring vignettes, which are supposed to account for DIF. Consequently, it is unlikely that the issue could be solved by increasing the number of experts, or perhaps crowdsourcing (Maestas, Buttice, and Stone Reference Maestas, Buttice and Stone2014): if all specialists on Western Europe or West European citizens have a systematically different understanding of electoral integrity than experts of other regions, errors are not random and will not cancel out by increasing the pool size. Due to this systematic bias on answers to each individual question, using other measures of central tendency (as suggested by Lindstädt, Proksch, and Slapin Reference Lindstädt, Proksch and Slapin2018) would still yield incomparable estimates across regions. Beyond incomparable averages, lack of loadings’ invariance also compromises these dimensions’ use in regression analysis. For instance, the solutions proposed by Marquardt and Pemstein (Reference Marquardt and Pemstein2018) can help correcting for scalar invariance and yield cross-regionally comparable country means. They might not, however, solve the lack of metric invariance for linear models and the biased regression estimates that follow. The only solution, given available data, would be to use Bayesian multilevel structural equation models with random factor loadings (Muthén and Asparouhov Reference Muthén and Asparouhov2018; Castanho Silva, Bosancianu, and Littvay Reference Castanho Silva, Bosancianu and Littvay2019). Such methods can model the cross-regional noninvariance and should be applied by researchers who use the electoral integrity dimensions in their regression analyses.

There might be a few ways to alleviate noninvariance problems when designing expert surveys. First, following PEI’s exemplary lead and having (a) constructs with a strong theoretical foundation, (b) multiple questions per construct, and (c) anchoring vignettes. This allows, at least, to test whether the measurement is cross-regionally valid. Second, one could also collect pilot data with quasi-experts (e.g., internationally diverse groups of social science graduate students) who are likely to respond similarly to the targeted respondents. Following the analytical steps presented here, one can identify, adjust or eliminate problematic items, and, if needed, even drop entire regions. We call for better design of expert survey batteries and more rigorous pretests, leading to comparative measurement of relevant concepts where results can actually be compared.

Supplementary material

For supplementary material accompanying this paper, please visit


Authors’ note: We owe a debt of gratitude to the PEI team, especially Thomas Wynter and Kristina Gushchina who put us in contact. Wynter and his team’s responsiveness to our inquiries were exemplary, he explicitly encouraged us to research language effects and worked with us to get what we needed. We are also thankful for comments on earlier versions of this manuscript by the three anonymous reviewers and Jeff Gill. The replication files for this article are available on the Political Analysis Dataverse (Castanho Silva and Littvay 2019).

Contributing Editor: Jeff Gill

1 Version 5.5, October 2017.

2 More details on sampling and methods in the Online Appendix.

3 Descriptive statistics in the Online Appendix, Table S1.

4 All items are in Table 2. Replication data and codes for all analysis in this paper are available on the Political Analysis Dataverse (Castanho Silva and Littvay Reference Castanho Silva and Levente2019).

5 Measurement (non)invariance is conceptually the same as Differential Item Functioning (DIF, Kline Reference Kline2016, p. 398). The main difference is that MI assumes a linear relationship between latent construct and indicators, treating indicators as continuous. DIF assumes a nonlinear relationship, treating variables as ordinal—for empirical and mathematical comparisons, see Muthén and Christoffersson (Reference Muthén and Christoffersson1981), and Raju, Laffitte, and Byrne (Reference Raju, Laffitte and Byrne2002). The PEI aggregation assumes continuous measurement, by summing up and standardizing responses (Norris, Wynter, and Grömping Reference Norris, Wynter and Grömping2017), in common with most expert surveys, so we apply Structural Equation Modeling based MI tests.

6 For a deeper discussion of advantages of different invariance tests, see Davidov, Muthén, and Schmidt (Reference Davidov, Muthén and Schmidt2018).

7 Vignettes changed in 2017, so we only use pre-2017 data in these tests. Technical details of the vignettes and their incorporation, and analyses without vignettes, in the Online Appendix.

8 Results with scalar invariance in the Online Appendix.

9 In the Online Appendix we show other ways beyond DIF-corrected indicators to incorporate vignettes. None produces more invariant results.


Asparouhov, T., and Muthén, B.. 2014. “Multiple-group Factor Analysis Alignment.” Structural Equation Modeling: A Multidisciplinary Journal 21(4):495508.CrossRefGoogle Scholar
Bakker, R., Edwards, E., Jolly, S., Polk, J., Rovny, J., and Steenbergen, M.. 2014. “Anchoring the Experts: Using Vignettes to Compare Party Ideology Across Countries.” Research & Politics 1(3):18.CrossRefGoogle Scholar
Castanho Silva, B., Bosancianu, C. M., and Littvay, L.. 2019. Multilevel Structural Equation Modeling . Thousand Oaks, CA: Sage.Google Scholar
Castanho Silva, B., and Levente, L.. 2019. “Replication Data for: Comparative Research is Harder than We Thought: Regional Differences in Experts’ Understanding of Electoral Integrity Questions.”, Harvard Dataverse, V1.CrossRefGoogle Scholar
Davidov, E., Muthén, B., and Schmidt, P.. 2018. “Measurement Invariance in Cross-National Studies: Challenging Traditional Approaches and Evaluating New Ones.” Sociological Methods & Research 47(4):631636.CrossRefGoogle Scholar
von Davier, M., Shin, H.-J., Khorramdel, L., and Stankov, L.. 2017. “The Effects of Vignette Scoring on Reliability and Validity of Self-Reports.” Applied Psychological Measurement 42(4):291306.CrossRefGoogle ScholarPubMed
Hu, L.-t., and Bentler, P. M.. 1999. “Cutoff Criteria for Fit Indexes in Covariance Structure Analysis: Conventional Criteria Versus New Alternatives.” Structural Equation Modeling: A Multidisciplinary Journal 6:155.CrossRefGoogle Scholar
Jöreskog, K. G. 1971. “Simultaneous Factor Analysis in Several Populations.” Psychometrika 36:409426.CrossRefGoogle Scholar
King, G., Murray, C. J. L., Salomon, J. A., and Tandon, A.. 2004. “Enhancing the Validity and Cross-Cultural Comparability of Measurement in Survey Research.” American Political Science Review 98(1):191207.CrossRefGoogle Scholar
Kline, R. B. 2016. Principles and Practice of Structural Equation Modeling . 4th edn. New York: The Guilford Press.Google Scholar
Lindstädt, R., Proksch, S.-O., and Slapin, J. B.. 2018. “When Experts Disagree: Response Aggregation and its Consequences in Expert Surveys.” Political Science Research and Methods, Scholar
Maestas, C. D., Buttice, M. K., and Stone, W. J.. 2014. “Extracting Wisdom from Experts and Small Crowds: Strategies for Improving Informant-based Measures of Political Concepts.” Political Analysis 22(3):354373.CrossRefGoogle Scholar
Marquardt, K. L., and Pemstein, D.. 2018. “IRT Models for Expert-Coded Panel Data.” Political Analysis 26(4):431456.CrossRefGoogle Scholar
Millsap, R. E. 2011. Statistical Approaches to Measurement Invariance . New York: Routledge.Google Scholar
Muthén, B. O., and Christoffersson, A.. 1981. “Simultaneous Factor Analysis of Dichotomous Variables in Several Groups.” Psychometrika 46(4):407419.CrossRefGoogle Scholar
Muthén, B. O., and Asparouhov, T.. 2018. “Recent Methods for the Study of Measurement Invariance With Many Groups: Alignment and Random Effects.” Sociological Methods & Research 47(4):637664.CrossRefGoogle Scholar
Norris, P., Wynter, T., and Grömping, M.. 2017. Codebook – The Expert Survey of Perceptions of Electoral Integrity, Release 5.5, (PEI_5.5). The University of Sidney: The Electoral Integrity Project.Google Scholar
Oberski, D. L. 2014. “Evaluating Sensitivity of Parameters of Interest to Measurement Invariance in Latent Variable Models.” Political Analysis 22(1):4560.CrossRefGoogle Scholar
Raju, N. S., Laffitte, L. J., and Byrne, B. M.. 2002. “Measurement Equivalence: A Comparison of Methods Based on Confirmatory Factor Analysis and Item Response Theory.” Journal of Applied Psychology 87(3):517529.CrossRefGoogle ScholarPubMed
Satorra, A., and Bentler, P. M.. 2001. “A Scaled Difference chi-Square Test Statistic for Moment Structure Analysis.” Psychometrika 66(4):507514.CrossRefGoogle Scholar

Castanho Silva and Littvay supplementary material

Castanho Silva and Littvay supplementary material 1

[Opens in a new window]
File 162 KB

Altmetric attention score

Full text views

Full text views reflects PDF downloads, PDFs sent to Google Drive, Dropbox and Kindle and HTML full text views.

Total number of HTML views: 193
Total number of PDF views: 463 *
View data table for this chart

* Views captured on Cambridge Core between 03rd July 2019 - 4th December 2020. This data will be updated every 24 hours.

Hostname: page-component-b4dcdd7-lltvg Total loading time: 1.258 Render date: 2020-12-04T18:35:00.055Z Query parameters: { "hasAccess": "1", "openAccess": "0", "isLogged": "0", "lang": "en" } Feature Flags last update: Fri Dec 04 2020 18:00:25 GMT+0000 (Coordinated Universal Time) Feature Flags: { "metrics": true, "metricsAbstractViews": false, "peerReview": true, "crossMark": true, "comments": true, "relatedCommentaries": true, "subject": true, "clr": false, "languageSwitch": true }

Send article to Kindle

To send this article to your Kindle, first ensure is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about sending to your Kindle. Find out more about sending to your Kindle.

Note you can select to send to either the or variations. ‘’ emails are free but can only be sent to your device when it is connected to wi-fi. ‘’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

Find out more about the Kindle Personal Document Service.

Comparative Research is Harder Than We Thought: Regional Differences in Experts’ Understanding of Electoral Integrity Questions
Available formats

Send article to Dropbox

To send this article to your Dropbox account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your <service> account. Find out more about sending content to Dropbox.

Comparative Research is Harder Than We Thought: Regional Differences in Experts’ Understanding of Electoral Integrity Questions
Available formats

Send article to Google Drive

To send this article to your Google Drive account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your <service> account. Find out more about sending content to Google Drive.

Comparative Research is Harder Than We Thought: Regional Differences in Experts’ Understanding of Electoral Integrity Questions
Available formats

Reply to: Submit a response

Your details

Conflicting interests

Do you have any conflicting interests? *