RESEARCH DISCIPLINE
Statistical Analysis for Social Psychology Research
Consulting for Experimental & Survey Studies
Social psychology statistical analysis for online and laboratory experiments, moderation and mediation studies, attitude surveys and dyadic data, with the primary outcome, exclusions and power settled before the data are examined and results prepared for your journal.
No Commitment Required
Strictly Confidential
In-House PhD-Level Statisticians
Fixed Project Fee
Same-Day Response
Statistical Analysis for Social Psychology Research
Social psychology statistical analysis is the statistical design, analysis and reporting of studies of how people think, feel and act in response to other people and social settings, using online and laboratory experiments, surveys of attitudes and identity, and dyadic or group data. Effects are usually small, manipulations are brief, many outcomes and moderators can be tested, and choices about covariates and exclusions are flexible, so a workflow that reports whichever contrast reaches significance, or claims an interaction by comparing separate p-values, yields findings that do not hold up. OnlineSPSS, a specialist statistical consultancy for academic, clinical and professional research, provides social psychology data analysis through in-house PhD-level statisticians, who settle the primary outcome, the smallest effect of interest and the exclusion rules with you before the data are examined.
WHEN TO ENGAGE
Statistical Support at Different Stages of Social Psychology Research
Researchers usually contact us at a recognizable point, from the design of a new experiment to a reviewer's challenge to an interaction effect.
Designing an Experiment You are planning an online or laboratory experiment and need the primary outcome, the smallest effect of interest, the sample that effect requires and the exclusion rules settled before data collection.
Analyzing Experimental Data Your data are collected, and the treatment of several outcomes, the manipulation check and the tests of moderators and mediators need to be chosen, run and checked against the plan.
Testing Moderation or Mediation You want to show for whom or why an effect occurs, and the interaction, the indirect effect and the wording that the design supports need to be worked out with their intervals.
Reviewer Questions on Power and Exclusions A reviewer or advisor has asked about power, flexible exclusions, unreported outcomes or the way an interaction was tested, and the analysis needs a clear and honest response.
RESEARCH CONTEXT
Research Designs and Data We Work With in Social Psychology
The datasets behind most social psychology data analysis come from short manipulations and self-report outcomes, and the way people were grouped decides the model.
01
Online and Survey Experiments
Participants randomized to vignettes or messages in an online panel, with attention checks, brief manipulations and outcomes on a few items.
02
Laboratory Interaction Studies
Participants interacting in pairs or groups, where the responses within a pair or group depend on one another.
03
Moderation and Mediation Studies
Experiments and surveys testing why an effect occurs or for whom, with a mediator or moderator measured in the same session.
04
Attitude and Intergroup Surveys
Cross-sectional and panel surveys of attitudes, identity and prejudice, with scale scores and sometimes implicit measures alongside them.
05
Dyadic and Relationship Data
Couples, friends or coworkers measured together, where each person's score is related to the partner's and the pair is the unit.
06
Replication and Multi-Site Studies
Direct replications and studies run across several laboratories, with the site as a clustering level and a prespecified smallest effect of interest.
METHODOLOGICAL CONSIDERATIONS
Statistical Issues We Consider in Social Psychology Research
Before any effect is reported, your power, your analytic choices and your handling of several outcomes are examined, because each can shift a finding more than the choice of test.
Small Effects and Power Published effect sizes in the area tend to be inflated, so power is judged against the smallest effect worth detecting, and a sensitivity analysis shows which effects the final sample could detect.
Flexible Analytic Choices Covariates, exclusions and outcome definitions are fixed in the plan. Where a choice is arbitrary, results under reasonable alternatives are shown, so readers can see how far the conclusion depends on it.
Several Outcomes and Checks Many outcomes and manipulation checks inflate the error rate if only the significant ones are reported. A primary outcome is named, the others are treated as a family, and all of them are reported.
Interactions Tested Directly A significant effect in one group and a non-significant one in another does not show a difference. The interaction term is tested with its interval, and simple slopes are interpreted after it.
Exclusions After the Manipulation Excluding participants who fail a check taken after the manipulation can unbalance the conditions. Exclusion rates by condition are reported, and results with and without exclusions are compared.
Partners and Groups Are Not Independent People in a pair, a group or a site resemble one another, so the model includes that level, and dyadic models estimate each person's effect on the other.
STATISTICAL METHODS
Methods Selected for the Research Question and Study Design
Each method family follows the design and the claim of your study, and the line under each title says why it suits social psychological data.
Regression & Predictive Modeling
Moderation is tested through the interaction term with simple slopes and intervals, and mediation and conditional indirect effects are described as associations unless the design supports more.
Comparative & Hypothesis Testing
Randomized conditions are compared as a difference with its interval, with covariates measured before the manipulation and rank-based tests for short ordinal outcomes.
Longitudinal & Multilevel Analysis
Participants in pairs, groups or sites share an environment, and attitudes in panel waves are correlated over time, so mixed and cross-lagged models carry both structures.
Structural & Latent Variable Modeling
Attitude and identity scales are tested for structure and for invariance across groups, and latent variables keep measurement error from weakening the estimated paths.
Bayesian & Computational Methods
Dropout after randomization in online studies is handled by imputation or sensitivity analysis, and bootstrapped intervals are used for indirect effects.
Evidence Synthesis & Meta-Research
A literature of small studies is combined with heterogeneity and checks for small-study effects, since a single study rarely settles a small effect.
DELIVERABLES
Statistical Outputs Prepared for Your Research
The outputs follow the structure of a social psychology study, from the review of the plan and exclusions to the scripts that regenerate every reported value.
Data & Study Review The preregistration or plan, the condition codes, the attention-check and exclusion flags, and the measures with their scoring are reviewed before any model is fitted.
Statistical Analysis The agreed condition, moderation and mediation analyses, with assumption diagnostics, correction for the family of secondary outcomes and the planned sensitivity analyses.
Effect Estimates & Uncertainty Condition effects and interactions with confidence intervals, standardized effects with their scaling stated, indirect effects with bootstrap intervals and a sensitivity power analysis.
Tables & Figures Descriptive tables by condition, effect tables, interaction plots with simple slopes, exclusion tables by condition and the participant flow from recruitment to analysis.
Statistical Interpretation Results are explained for the theory under test, confirmatory and exploratory findings are marked apart, and the limits of the sample, the setting and the measures are stated.
Analysis Files R scripts, SPSS syntax or Mplus input files, with condition coding and exclusion rules documented so that each reported value can be regenerated.
REPORTING
Reporting Social Psychology Research
The expectations depend on the journal and on whether the study was registered, and each is applied only where it fits.
Studies are prepared to APA JARS-Quant, which asks for the design, the manipulation and its checks, the rationale for the sample size, the exclusions and the effect estimates with intervals. Indirect effects are reported to AGReMA, with the assumption that no unmeasured variable affects both mediator and outcome stated beside the estimate. Where a journal follows the TOP Guidelines, the preregistration, data and code are cited as those guidelines specify. Each departure from a registered plan is reported, exclusions are tabulated by condition and every outcome that was measured is accounted for, so the manuscript and the output agree.
ILLUSTRATIVE EXAMPLE
How the Analysis May Look in Practice
An illustration of how an online experiment with a moderator moves from a plan to a defensible, reportable result.
ILLUSTRATIVE EXAMPLE · INTERGROUP CONTACT RESEARCH · UNITED STATES
A social psychology group tested whether an imagined-contact message reduced prejudice toward an outgroup in adults recruited online, and whether the effect depended on prior contact.
The study had one attitude scale as its primary outcome, two further outcomes and an attention check taken after the message. Prior contact was measured before randomization, so the moderator could not be affected by the message. Before the data were examined, we fixed the plan in writing: the primary outcome, correction across the two further outcomes as a family, and an interaction term for prior contact with simple slopes after it, not separate subgroup tests. Exclusion rates by condition were reported and all results were repeated without exclusions. A sensitivity power analysis showed which effect sizes the sample could detect. A mediator measured after the message was treated as an association, because it was not randomized.
Outcome: The group received the condition effect and the interaction with 95% confidence intervals, the sensitivity power analysis, exclusion tables by condition, a methods paragraph 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.
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.
SOFTWARE
Statistical Software for Social Psychology Research
Social psychology analyses are run in the software your laboratory already uses, with syntax or scripts provided where applicable.
Statistical Support Within Your Research OnlineSPSS provides the agreed methodological, analytical, and statistical reporting support. The researcher or research team retains the scientific question, substantive interpretation, conclusions, and overall responsibility for the research.
FREQUENTLY ASKED QUESTIONS
Questions About Statistical Analysis for Social Psychology
Short answers to what researchers ask before sharing a study, from interactions and several outcomes to sample size, confidentiality and the assessment.
What statistical methods are used in social psychology research?
Social psychology most often uses comparisons of randomized conditions, moderation and mediation analysis, multilevel and dyadic models for people in pairs or groups, and latent variable models for attitude scales, reported with effect sizes and confidence intervals. The choice depends on the design, on how participants were grouped and on whether the study was registered. Our Statistical Methods pages show how each family is applied.
How should I test whether an effect differs between groups?
Whether an effect differs between groups is answered by testing the interaction directly, with its confidence interval, and not by finding that one simple effect is significant and the other is not. The interaction term compares the effects, and simple slopes are then interpreted to describe where the effect is present and how large it is.
How do you handle several outcomes and manipulation checks?
A primary outcome is named in the plan, and the remaining outcomes are treated as a family with an error-rate correction such as Holm. All outcomes that were measured are reported. A manipulation check is reported for every participant, and it is used as an exclusion criterion only if that rule was set in advance and its rates are compared across conditions.
How do I estimate the sample size for a small effect?
We start from the smallest effect that would be of interest, not from a single published estimate, and show how the required sample changes with that choice. Expected dropout after randomization and the exclusions allowed by the plan are added to the target. The calculation is written into the analysis plan.
How is confidential research data handled?
Study datasets, protocols and analysis files are handled confidentially, and project information is not published. 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, 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
Discuss Your
Social Psychology Research With Us
Share your research question, design and available data, even if only a draft plan or pilot data exist. We review how participants are grouped and confirm the approach before any work begins.
No Commitment Required
Strictly Confidential
Fixed Project Fee
In-House PhD-Level Statisticians
Same-Day Response
Serving Researchers Since 2014