RESEARCH DISCIPLINE
Sociology Research Project Statistics
Measuring Inequality Between Groups
For social inequality data analysis in income, schooling, health or housing: gaps expressed on a scale that suits the outcome, survey design respected throughout, and group comparisons reported without misleading the reader.
No Commitment Required
Strictly Confidential
In-House PhD-Level Statisticians
Fixed Project Fee
Same-Day Response
What Does Statistical Work on Inequality Outcomes Involve?
Sociology research project statistics concerns outcomes that differ between social groups, such as income, schooling, health or housing, and the surveys in which they are measured. It covers deciding how a gap is expressed, whether as a difference, a ratio or a difference in probabilities, how far to adjust for other characteristics, and how to compare groups when the outcome is skewed or binary. OnlineSPSS, a specialist statistical consultancy for academic, clinical and professional research, supports sociologists and survey researchers: an in-house PhD-level statistician takes on the modeling of group gaps in your survey and states what a gap between groups does and does not establish. Why the gaps arise is a question your theory answers, not the model.
WHEN TO ENGAGE
Decisions That Shape an Inequality Study
The way a gap is measured is settled before any model is run, and every later stage inherits that decision whether or not it was made deliberately.
Outcome and Gap Defined The outcome is skewed income, a yes-or-no event or an ordered category, and the form of the gap, a difference, a ratio or a change in probability, is fixed so that the question has a clear answer.
Groups to Compare Categories of race, gender, class or region are being coded, and the number of respondents in each decides how precisely any gap between groups can be estimated.
Intersections to Examine Social stratification models often concern combinations such as gender within class or race within region, and the plan decides between interaction terms and separate groups, knowing that each split thins the sample.
Findings Under Scrutiny A reviewer or committee asks whether coefficients can be compared across groups, or whether an adjusted gap still supports the claim made in the abstract.
WHAT THE SUPPORT COVERS
Measuring and Explaining Group Gaps
Inequality research turns on how a gap is measured and how far it is explained, and both are handled with the survey design kept in view.
Measuring the Outcome
Skewed income is analyzed on a log scale or by quantiles, and wealth, which can be zero or negative, by quantiles or a transformation that allows for it, so that a gap appears at several points of the distribution, not only at the mean.
Defining Group Gaps
Gaps are stated as absolute and relative differences side by side, since the same relative gap can mean very different absolute gaps when an outcome is rare or common.
Binary Outcomes
Differences in a yes-or-no outcome are analyzed with Logistic Regression and reported as predicted probabilities, because coefficients from separate group models cannot be compared directly.
Decomposing Inequality
An Oaxaca-Blinder decomposition divides a gap in a continuous outcome into a part linked to characteristics and a part linked to returns to them, with the reference group stated. A Gini and Theil decomposition splits overall inequality into between-group and within-group parts: the Theil index separates them exactly, while the Gini adds an overlap term, which is reported.
Regions and Places
Sociological survey modeling of respondents within regions uses Linear Mixed Models (Multilevel Modeling) to show how much of a gap lies between places, with the number of places stated as a limit.
Reporting Inequality Findings
Gaps are reported with confidence intervals and the group sizes behind them, as sociology journals expect of Sociology research and of Education & Social Sciences more generally.
COMMON ISSUES
Distortions in Estimates of Inequality
A gap between groups is easy to compute and easy to misstate, and each problem below changes either its size or its meaning.
Raw Mean Income Compared Mean income is compared across groups although a few very high earners dominate it, so the gap shifts with the top of the distribution and with any truncation of top values in the file.
Ratio Without Difference A twofold difference in a rare outcome is reported without the absolute difference, and readers are left to judge importance from the ratio alone.
Controlling the Gap Away Education, occupation and income are all added to a model of a group gap, though they may be the route through which the difference arises, and the small remaining gap is read as no inequality.
Coefficients Compared Across Groups Logit coefficients from separate models for two groups are compared as if on one scale, though a difference in unobserved variation alone can change them.
Small Subgroups Over-Read A gap between two small subgroups is reported as a finding with its wide interval left out, so a difference that sampling variation could explain is treated as established.
Broad Categories Hide Variation Wide group labels merge populations whose circumstances differ, and the gap for the merged group may describe none of its parts.
DELIVERABLES
Prepared Materials for an Inequality Paper
The materials fix how each gap was measured, so that a reader can see the scale, the adjustment and the group sizes behind every number.
Gap Definition Table Each gap with its scale, its reference group and its form, absolute or relative, set out before the models are shown.
Weighted Descriptive Tables Group means, medians and proportions that use the survey weights, with the number of respondents in each group and 95% confidence intervals.
Predicted Outcome Plots Plots of predicted outcomes by group with their intervals, so that differences are read in the units of the outcome and not as raw coefficients.
Adjustment Sequence Table Estimates of each gap before and after each block of covariates, with the reasoning for the order in which they enter the model.
Between-Place Variance Summary How far each gap differs from one region to another, with the number of regions stated.
Recoding Record and Code A record of how income, education and group variables were recoded, with the scripts that regenerate every table and figure.
ILLUSTRATIVE EXAMPLE
Completion Gaps in a National Youth Survey
An illustration of how a gap in a yes-or-no outcome is analyzed when adjustment could remove part of the difference being studied.
ILLUSTRATIVE EXAMPLE · EDUCATIONAL MOBILITY RESEARCH · UNITED STATES
A doctoral candidate in sociology wanted to know how much of the difference in college completion between young adults from lower-income and higher-income families remained after accounting for earlier school achievement, using a national longitudinal survey.
The plan set out the gap as a difference in predicted probability, the covariates and the order in which they would enter before any result was inspected. The complication was that school achievement may itself be one route by which family income shapes completion, so adjusting for it could understate the total gap. Both the unadjusted gap and the gap after each block of covariates were reported, survey weights were applied, and group sizes were shown beside each estimate. The finding was described as an association between family income and completion, with the reasoning about pathways explained.
Outcome: The candidate received the predicted probabilities of completion by family income with their 95% confidence intervals, the adjustment sequence table, the weighted descriptive tables and the scripts.
This example does not describe a real client or project. We never share details of actual client work, in line with the confidentiality we promise every researcher. This illustrative example is included to show the type, scope, and standard of work involved in this kind of analysis.
SOFTWARE
Survey Software for Group Comparisons
Weights, strata and clusters must be declared once and carried through every estimate, so the choice of package follows what your department teaches.
HOW WE ASSESS YOUR PROJECT
We Review Your Actual Research
Your assessment is based on a review of your actual research materials and analytical requirements, not a generic project category.
You Submit Your Project Details
Share whatever is available at the current stage, such as the research aim, protocol or proposal, dataset, variable list, manuscript or statistical requirements.
We Review Your Research and Define the Scope
Your statistically trained Research Coordinator reviews the materials to understand the study design, research questions, data structure, analytical requirements and expected deliverables. If clarification is needed, additional documents or information can be requested before the assessment is finalized.
Your Analysis Is Completed and Delivered
Once the scope and fixed project fee are agreed, the in-house statistician completes the work and the Research Coordinator remains the communication point through delivery, with the analysis, deliverables and timeline as agreed.
RESEARCH RESPONSIBILITY
Interpreting the Gaps Is Your Work
OnlineSPSS supports the statistical measurement and modeling of inequality, and the theory, the interpretation of the gaps and the conclusions are the researcher's.
Agreed Statistical Support OnlineSPSS provides the agreed methodological, analytical and statistical reporting support, defined in a scope that is confirmed before any work begins.
No Intellectual Contribution OnlineSPSS does not claim any intellectual contribution to your research. Authorship of the dissertation, paper or report remains yours.
Your Research Decisions The research question, the theoretical framing, the sociological interpretation, the discussion and the conclusions remain yours, with overall responsibility.
Confidential Handling Research data and project details are handled confidentially, project information is not published, and an NDA is available where required.
FREQUENTLY ASKED QUESTIONS
Questions About Studying Inequality With Survey Data
Inequality estimates invite technical objections from reviewers, three of which are met directly below.
What is covered when inequality outcomes are analyzed in survey data?
Social inequality data analysis covers the measurement and modeling of group differences in survey data: how a gap is defined, the model suited to the outcome, adjustment, survey design and the reporting of gaps with their uncertainty. It suits dissertations, papers and institute reports on inequality.
Can odds ratios be compared between two groups?
Not directly when they come from separate models. A logit coefficient reflects unobserved variation in each group as well as the association, so a difference may not mean what it appears to. Predicted probabilities and average marginal effects, from the separate models or from a pooled model with group interactions, put the comparison on the probability scale, which both groups share.
Should I adjust for education when studying an income gap?
It depends on the question. If education is part of how the gap arises, adjusting removes that portion, and the remaining gap answers a narrower question. Reporting the gap before and after adjustment, with the reasoning for each covariate, lets a reader see both the total difference and what remains.
How should skewed income be analyzed?
A log transformation expresses differences as percentage differences and reduces the pull of high values, though zero incomes need a decision. Medians and quantile models describe the typical value or particular points of the distribution, and top-coded values are noted. The choice depends on whether the question concerns the average, the typical or the top.
How do you protect my research data and confidentiality?
Your data and project details are handled confidentially, and project information is not published. Researcher names and institutions are not used publicly without permission. An NDA is available where required and can be arranged before sensitive materials are shared. Read more about confidentiality and data security.
What happens when I request an assessment, and how long does it take?
You share your research question, design and relevant materials, such as a proposal, the survey codebook or draft models, and a statistically trained Research Coordinator reviews the request and responds the same day. The analysis, deliverables, timeline and one fixed project fee are then confirmed before any work begins, so the timeline follows from the agreed scope.
Still have a question? Find more answers in our full FAQ.
NEXT STEP
Studying a Gap
Between Social Groups?
An inequality finding is only as clear as the scale chosen for the gap, which is why that choice is made first and stated in every table.
No Commitment Required
Strictly Confidential
Fixed Project Fee
In-House PhD-Level Statisticians
Same-Day Response
Serving Researchers Since 2014