RESEARCH DISCIPLINE
Statistical Analysis for Strength & Conditioning Research
Consulting for Athlete Testing & Training Studies
Strength and conditioning statistical analysis for jump, sprint and strength test batteries, training program comparisons and season-long monitoring, with reliability, the smallest worthwhile change and multiplicity handled so that the results stand up in review.
No Commitment Required
Strictly Confidential
In-House PhD-Level Statisticians
Fixed Project Fee
Same-Day Response
Statistical Analysis for Strength & Conditioning Research
Strength and conditioning statistical analysis is the statistical design, analysis and reporting of studies of athletic testing and training, from jump, sprint and strength test batteries to comparisons of training programs across blocks of a season. Squads are small, the same athletes are tested repeatedly, each test carries its own measurement error and a battery yields many metrics and time points, so a workflow that reads every change as real, judges reliability by a correlation or tests every metric against zero produces findings that do not hold. OnlineSPSS, a specialist statistical consultancy for academic, clinical and professional research, provides sports performance data analysis through in-house PhD-level statisticians, who settle the reliability of each test, the smallest worthwhile change and the primary outcomes with you before any comparison is made.
WHEN TO ENGAGE
Statistical Support at Different Stages of Strength & Conditioning Research
Support is most often requested when a test must be shown fit for use, a program must be compared fairly or a season of data must answer a question.
Validating a Test Battery You are introducing or comparing jump, sprint or strength tests and need reliability, typical error and the smallest worthwhile change established before they are used.
Comparing Training Programs You are comparing two training approaches across blocks of a season and need a design and a model that suit small squads, repeated tests and athletes who miss sessions.
Analyzing a Season of Monitoring Data You hold a season of test results, training load or wellness records for a squad, and need the repeated structure, the gaps and the many metrics handled.
Defending a Result Before Review A reviewer or editor has questioned the reliability statistic, the size of the sample or the number of metrics tested, and the reply needs a corrected analysis.
RESEARCH CONTEXT
Research Designs and Data We Work With in Strength & Conditioning
Athlete testing statistics start from squads of modest size tested again and again, and what each study records, tests, loads or seasons, decides which analysis fits.
01
Test-Retest Reliability Studies
Athletes tested twice or more under the same conditions on separate days, so that the typical error and consistency of a test can be estimated before use.
02
Jump, Sprint and Strength Batteries
Several tests per athlete in a session, such as a countermovement jump, a sprint and a repetition maximum, each with several attempts and a best or mean value.
03
Training Intervention Studies
Two or more programs compared across weeks or blocks, with tests at the start and end, athletes who miss sessions and sometimes allocation by whole squad.
04
Season-Long Monitoring Data
Repeated tests, training load and wellness scores for each athlete across a season, recorded at uneven intervals with gaps from injury and travel.
05
Velocity and Load Profiling
Bar velocity, power and load recorded across several loads, from which individual load-velocity relationships and maximum estimates are derived.
06
Squad and Position Comparisons
Athletes from different squads, age groups or positions compared on the same tests, where small and unequal groups limit what can be claimed.
METHODOLOGICAL CONSIDERATIONS
Statistical Issues We Consider in Strength & Conditioning Research
A change in a test score means little until the noise of that test and the size of the sample are known, so both are examined before any comparison.
ICC, CV and Typical Error Performance test reliability analysis separates the questions: the ICC shows consistency against between-athlete spread, while typical error and the CV express the test's noise in its unit or in percent.
Correlation Is Not Agreement Two sets of scores can correlate strongly and still differ by a constant or proportional bias. Agreement is shown with an ICC form that matches the design and with limits of agreement.
Smallest Worthwhile Change A change is judged against both the typical error and a smallest worthwhile change defined before the data are seen, and the interval of the change is reported beside them.
Many Tests, Metrics and Time Points A battery of tests, metrics and time points produces many comparisons. A primary outcome, a prespecified family and an adjustment keep a chance finding from being reported as an effect.
One Athlete Can Move the Mean With few athletes, intervals are wide and a single extreme result can shift the squad average. Influential athletes are inspected, and conclusions are limited to the precision the sample gives.
Regression to the Mean Athletes picked because of a low or high test score drift toward the average on retest, so a training effect needs a comparison group selected the same way, with the analysis adjusted for baseline.
STATISTICAL METHODS
Methods Selected for the Research Question and Study Design
Each method family is chosen for the outcome and the data structure of your study, and the line under each title explains why it is used in strength training research statistics.
Multivariate, Scale & Reliability
Test-retest data are summarized with the ICC, typical error and CV, and correlation analysis is kept for association between tests, not for agreement.
Longitudinal & Multilevel Analysis
A season of repeated tests is modeled with athlete effects and, where squads differ, a squad level, so an athlete who misses a test is kept, and counts or binary outcomes use generalized models.
Comparative & Hypothesis Testing
Programs are compared with a baseline-adjusted analysis, usually more precise than a comparison of gain scores; a mixed ANOVA suits tests repeated across several blocks; and paired comparisons describe change within one program.
Regression & Predictive Modeling
Load-velocity and dose-response relationships, and links between test results and performance, are modeled with few predictors and reported with intervals.
Bayesian & Computational Methods
With squads this small, Bayesian estimation with stated priors can express uncertainty about the smallest worthwhile change, and bootstrap intervals are checked against the model-based ones.
Evidence Synthesis & Meta-Research
Strength training trials are small, so a meta-analysis or a meta-regression on dose, such as weekly volume, summarizes them with heterogeneity shown.
DELIVERABLES
Statistical Outputs Prepared for Your Research
The outputs follow the structure of a testing and training study, from the review of the protocols to the scripts that regenerate every reported estimate.
Data & Study Review The test protocols, the number of attempts, how best or mean values were chosen, the interval between sessions, missed tests and the squad structure are reviewed before analysis.
Statistical Analysis Reliability analysis, baseline-adjusted or mixed models for training comparisons and the planned sensitivity analyses, including the effect of removing influential athletes.
Effect Estimates & Uncertainty The ICC with its interval, typical error and CV, and training effects as differences with confidence intervals, set against the smallest worthwhile change.
Tables & Figures A reliability table for each test, individual change plots with the typical error band, group effect plots and a table of every metric that was tested.
Statistical Interpretation Each result is explained in training terms, changes are read against measurement error and the limits of a small squad are stated for the discussion.
Analysis Files R or Python scripts, SPSS syntax or Stata programs, with the coding of tests, attempts and sessions documented so that every table can be regenerated.
REPORTING
Reporting Strength & Conditioning Research
The standards that apply depend on the study type, and each is used only where it fits.
A randomized training comparison reports to CONSORT 2025, with the extension for cluster randomized trials where whole squads are randomized, and the program is described to TIDieR so that load, progression and supervision can be repeated. Observational monitoring studies follow STROBE. Studies of test reliability and agreement report to GRRAS, which asks for the number of athletes and repeats, the form of the ICC and its confidence interval. Each reported test gives its typical error, so readers can judge a change against it, and the primary outcome and the family of tests are stated before the results. Where the target journal publishes statistical guidance for authors, the tables and methods text follow it.
ILLUSTRATIVE EXAMPLE
How the Analysis May Look in Practice
An illustration of how a small squad study moves from a test's noise to a program comparison stated with its limits.
ILLUSTRATIVE EXAMPLE · YOUTH ACADEMY FITNESS TESTING PROGRAM · UNITED STATES
A youth academy's sport science team wanted to know whether a revised strength program improved countermovement jump height more than the existing one over a preseason, and whether the test could detect it.
Two squads followed different programs, and the jump test was repeated on two baseline days. Before outcomes were examined, we fixed jump height as the primary outcome and estimated the typical error and the ICC from the repeated baseline tests. The smallest worthwhile change was defined in advance. The program difference was estimated with a model adjusted for baseline, so regression to the mean did not inflate the apparent gain. Because whole squads, not individual athletes, were allocated, the report states that squad differences cannot be separated from the program.
Outcome: The group received the typical error and ICC with their intervals, the baseline-adjusted program difference with its 95% confidence interval, individual change plots, 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.
RESEARCHER VOICE
What Researchers Say
Feedback from a performance scientist on support that understood the field, on a small sample with unstable estimates and on reporting framed for review.
“We needed statistical support from someone who understood Strength & Conditioning, not just someone who could run software. The main issue in our data was a relatively small sample and unstable estimates. The analysis and reporting were framed around the way studies in our field are actually designed and reviewed, which made a noticeable difference.”
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 Strength & Conditioning Research
Testing and training analyses are run in the package your team 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 Strength & Conditioning Research
Short answers to what strength and conditioning researchers ask before sharing a study, from reliability and worthwhile change to program comparisons and the assessment process.
Which reliability statistic should I report for a fitness test?
Report more than one, because each answers a different question. The ICC shows consistency relative to the spread between athletes and depends on how varied the sample is, while typical error and the coefficient of variation show the noise of the test in its own units or in percent. Each is given with a confidence interval, and a Pearson correlation alone is not enough.
How big a change counts as a real improvement?
A change counts when it exceeds the noise of the test and matters in practice. We compare it with the typical error and with a smallest worthwhile change, often set as a fraction of the between-athlete standard deviation, and justify that choice for your sport. The interval of the change is reported, not only whether it is positive.
Can I compare two training programs using two small squads?
You can, but the limits should be stated. If whole squads follow different programs, differences between squads cannot be separated from the effect of the program, so the result is worded with that caution. Adjusting for baseline accounts for differences in starting scores, though not for other ways the squads differ, and randomizing athletes or squads where possible strengthens the design.
Do I need a multiplicity correction for a test battery?
The first step is to name a primary outcome and a small family of main tests in advance. For those, an adjustment such as Holm’s method controls false positives, and the remaining tests are reported as exploratory with their intervals. Reporting only the significant tests misleads readers.
How is athlete testing data handled?
Test results, monitoring files 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 send your research question, the testing protocol and any results you hold. A statistically trained Research Coordinator reviews them and responds the same day, and the analysis, deliverables, timeline and one fixed project fee are confirmed before 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
Strength & Conditioning Research With Us
Share your research question, the testing protocol and the results you have, even if only a plan exists. We review how tests and sessions are structured and confirm the approach before work begins.
No Commitment Required
Strictly Confidential
Fixed Project Fee
In-House PhD-Level Statisticians
Same-Day Response
Serving Researchers Since 2014