Experimental Psychology Toolkit

Step 1

Upload a stimulus CSV

Start with the file you want to inspect. The app previews the data, guesses likely column types, and flags import problems before any statistics run.

Session ready

Clean stimulus file checklist

A predictable CSV makes the statistics easier to trust.

Use one row per stimulus and keep one stable ID column.
Put the word or phrase in one text column, for example word or stimulus.
Add a group or condition column when you want to compare matched sets.
Keep numeric ratings in numeric-only columns and avoid mixed text such as 4 high.
Export a clean spreadsheet file and keep the same separator throughout the file.

Included examples

Download a realistic or small CSV and upload it to see the workflow with known column names.

Word norming ratings word_norms_realistic.csv has valence, arousal, concreteness, imageability, and familiarity columns.

Download

Sentence stimuli sentence_stimuli_realistic.csv uses semicolons and decimal commas for cloze and plausibility ratings.

Download

2-group matching two_group.csv shows A/B stimulus matching with length, frequency, and valence.

Download

3-group ANOVA three_group.csv demonstrates A/B/C screening and pairwise checks.

Download

QC only no_group.csv includes duplicates, an empty stimulus, and a bad number.

Download

CSV upload Drop a CSV here or choose a file Comma, semicolon, and tab-delimited files are supported. The file stays in your temporary browser session.

Choose CSV No file selected

CSV import settings

Choose the delimiter before uploading. If the preview looks wrong after upload, change it here and re-read the same file without selecting it again.

CSV delimiter

Use Auto for most files. Choose a delimiter manually if your preview shows one giant column or values merged together.

Preview

No dataset loaded

Check whether the parser split the columns correctly and whether the first rows look like the original spreadsheet.

QC warnings

Early import checks

This block highlights file-shape problems such as width mismatches, renamed duplicate headers, or other issues seen during upload.

Step 2

Map columns and confirm the dataset

Tell the app which column contains the stimulus text, which column defines the comparison groups, and which columns should become numeric metrics.

Stimulus text column

Use the column containing the actual item shown to participants, such as a word, phrase, sentence, or picture label.

Group or condition column

Use this when comparing stimulus sets, for example neutral vs emotional, list A vs list B, or condition labels.

Stimulus ID column

Stable IDs make exports traceable after filtering, randomization, and importing into PsychoPy, jsPsych, Gorilla, or similar tools.

Numeric metrics

Uploaded quantitative columns

Choose the numeric columns that should be compared across groups, such as ratings, lengths, frequencies, or any precomputed item property.

For stronger studies, add validated or participant-rated columns such as frequency, valence, arousal, concreteness, familiarity, imageability, or rating SD.

Ignored columns

Keep them in exports but exclude from analysis

Use this for notes, labels, bookkeeping fields, or anything that should remain attached to each row but not drive the statistics.

Type overrides

Manual fixes when auto-detection is wrong

If a numeric column was read as text, or a note column was mistaken for data, correct it here before saving the mapping.

Column	Inferred type	Example	Override

Step 3

Run normalization analysis

Set the matching criteria here. The results table shows whether each metric passes the current screening rule and which rows look risky.

Waiting for dataset

Interpretation guardrails

Important for psychological study planning

A non-significant t-test or ANOVA does not prove that groups are equivalent. Treat "pass" as a planning screen, not a publication-ready equivalence claim.
Choose thresholds before looking at the results. Report alpha, outlier method, outlier cutoff, excluded rows, and the practical matching bounds you used.
If you need to claim equivalence scientifically, use an equivalence framework such as TOST with a justified smallest effect size of interest.
Small groups are unstable. The app disables inferential checks when any group has fewer than 3 usable observations for a metric.

Alpha

Alpha is the p-value cutoff used for the difference test. A p-value above alpha is not proof of equality, so also inspect effect sizes and thresholds.

Max standardized mean delta

This practical bound limits how far apart group means can be in standard deviation units. Smaller values force closer stimulus matching.

Max SD ratio

This compares the spread of the groups. A value near 1 means the groups vary by about the same amount.

Optional max raw mean delta

Use this when you want a practical limit in the original units, for example no more than 0.5 rating points or 1 character difference.

Outlier method

Z-score flags values far from the mean. IQR flags values far outside the middle range and is often safer when ratings are skewed.

Outlier cutoff

Lower cutoffs mark more rows as unusual. Higher cutoffs are stricter and only flag more extreme values.

Include derived text metrics Exclude flagged outliers from inferential tests Restrict analysis to rows without QC issues

Metric overview

Group means for the selected metric

Use the chart for a quick visual comparison only. The decision still comes from the metric table and its pass/fail checks.

Metric results

Pass/fail, p-values, thresholds

Each row summarizes one metric. Start with the status column, then inspect the test, effect size, and notes if a metric fails.

For manuscripts, report the descriptive statistics and effect sizes. Use this screen to detect confounds before running the experiment.

Row annotations

Inspect missing values, duplicates, parse failures, and outliers

This block explains why individual rows may be risky to keep, for example because a value is missing, duplicated, or unusually extreme.

Step 4

Generate counterbalanced lists and export files

After QC and analysis, create balanced list assignments and export the exact rows you want to carry into the experiment build.

Lists

Choose how many output stimulus lists you want to create for counterbalancing.

Seed

Keep the same seed if you want the exact same randomized assignment again later.

Stratify by

Stratification keeps assignments balanced within selected categories, such as valence class or condition label.

Block by

Blocking groups rows into separate assignment buckets before the lists are generated.

Keep groups separate during assignment

List balance

Assignments by group and list

Check that each list receives a balanced number of rows before exporting the randomized file.

Assignments

Per-row seeded output

The seed makes the assignment reproducible. Keep it if you want to rebuild the same lists later.

Research Tools

Participant Analysis Studio

Upload participant or trial-level CSV data, compare outcomes, review questionnaire items, and avoid common mistakes with repeated observations.

What this module does

Practical paper-writing checks

Compare outcomes between groups with beginner-safe defaults and effect sizes.
Check repeated rows so trial-level data are not analyzed as independent participants.
Use questionnaire item columns for reliability and exploratory factor analysis.
Use the notes to write cautious results sections without overclaiming.

Participant CSV checklist

Prepare the file before upload

Use one anonymous participant ID column. For trial-level files, repeat the same ID across that participant's rows.
Add a stimulus or item ID column when the same words, images, or trials are seen by many participants.
Keep group names consistent, for example control and treatment.
Store outcomes and questionnaire items as numeric-only columns.
Reverse-score items before upload when a scale requires it.
Avoid direct personal identifiers such as names, emails, or phone numbers.

Try the example CSV

Known columns for testing

Use the participant-level example for reliability/EFA and the long-format example for repeated observations with participant and stimulus IDs.

Participant-level CSV Long-format CSV

Suggested workflow

Upload, map, analyze, export

1. Upload: confirm the preview and delimiter.
2. Map: choose participant ID, optional stimulus ID, groups, conditions, outcomes, items, and covariates.
3. Analyze: inspect group comparisons, repeated-row checks, reliability, factor suggestions, and item loading patterns.
4. Export: download the analysis CSV for reporting or further review.

Participant CSV Upload participant data Use one row per participant or one row per observation. Include participant ID, optional item ID, groups/conditions, outcomes, and items.

Choose participant CSV No file selected

Import settings

CSV delimiter

Use Auto for most files. Choose comma, semicolon, or tab manually if the preview shows one large column or shifted rows.

Download example participant CSV

Participant preview

No dataset loaded

QC warnings

Import and design checks

Mapping

Define participant data columns

Choose the workflow first. If the same participant appears on several rows, map the participant ID before any interpretation.

Analysis mode

Choose the workflow

Mode

Use participant-level mode for one row per participant. Use repeated-observations mode when one participant has several trials, stimuli, or time points.

If the same participant appears on multiple rows, those rows are dependent. The repeated-observations helper gives a cautious approximate check for continuous outcomes, not a full advanced multilevel analysis.

Participant ID

Required for the repeated observations helper. Use an anonymous stable ID so trials or time points from the same participant are recognized as related.

Item / stimulus ID

Use this when the same stimuli appear for several participants. This helps separate stimulus differences from participant or group differences.

Group column

Use this for between-subjects groups such as control/treatment. Two groups use Welch's t-test; three or more use one-way ANOVA plus corrected pairwise checks.

Condition column

Use this for within-subject or trial condition labels, for example baseline vs primed or time 1 vs time 2.

Outcome columns

Group comparisons

Choose dependent variables such as accuracy, reaction time, score, or rating totals.

Questionnaire item columns

Reliability and EFA

Choose item-level variables, for example Likert questions q1 to q12.

Covariates

Optional numeric adjustment columns

Use for numeric covariates such as age or baseline score. Keep this small and theoretically justified for beginner analyses.

Repeated observations helper

Optional, continuous outcomes only

Use this only when rows repeat within participants. It gives a simple adjusted check for dependent rows; advanced multilevel models still belong in R, jamovi, JASP, SPSS, or consultation with a methods expert.

Outcome to check

Choose one continuous numeric outcome such as reaction time, rating, score, or accuracy proportion. Binary yes/no outcomes and counts are not supported here.

Repeated-row pattern

Start simple. Add stimulus/item adjustment only when the same stimuli are repeated across participants.

Analyze

Run participant checks

Alpha

The p-value cutoff for group tests. Keep effect sizes, sample size, and design assumptions visible; a p-value alone is not enough for interpretation.

Factor count

Use 0 to let parallel analysis suggest retention. Enter a number only when theory or a preregistered plan justifies it.

Rotation

Varimax is orthogonal and easy to read. If psychological factors are expected to correlate, treat this as exploratory and consider oblique rotation later.

Parallel iterations

Parallel analysis compares observed eigenvalues against random data. More iterations are slower but make the simulated 95% cutoff more stable.

Row-wise structure

Observation-level checks

Repeated observations result

Dependent-row check

Group tests

Welch t-test or ANOVA

Reliability

Alpha and item checks

Exploratory factor suggestions

Suitability and factor count

Item loading pattern

Exploratory item structure

Scientific notes

Interpretation guardrails

Planning

Study Planning Studio

Estimate sample size for common psychology study designs. These are assumption-based planning numbers, not guarantees.

Scenario-based planning

What this module supports

Beginner-friendly planning scenarios

Compare two independent groups.
Compare repeated measurements from the same participants.
Compare several independent groups.
Plan for a simple correlation.

Planning checklist

Before trusting the number

Use conservative effect sizes when prior evidence is uncertain.
Add expected exclusions, attrition, and unusable trials after the estimate.
Use achieved power only for planning a proposed sample size, not for explaining completed results.
Advanced repeated-measures and multilevel power are not implemented in this first version.

Planning question

Choose a study scenario

Study scenario

Choose the design that best matches your planned study. Keep it simple unless your supervisor or methods plan requires more.

Planning question

Most users should start with required sample size. The other options are useful when you already have a planned N.

Alpha Desired power Expected effect size

Use d for two groups or repeated measurements, f for several groups, and r for correlation. If unsure, use a conservative value from prior research.

Planned sample size Group ratio

Group 2 divided by group 1. Use 1 for balanced groups, 2 for twice as many in group 2, etc.

Number of groups Repeated-measure correlation

Expected correlation between paired measurements. Higher correlation usually reduces required participant count for paired comparisons.

Planning curve

How sample size changes power

Assumptions and limits

Read before using the estimate

Run a planning scenario to see assumptions and approximation notes.

Presentation

Results & Figures Studio

Build simple paper-ready figures and tables from CSV data or toolkit results. This workspace formats results; it does not add new statistical claims.

No figure data loaded

What this module does

Paper-ready presentation

Make bar charts, boxplots, scatter plots, and repeated-measure line plots.
Format mean ± SD tables and reuse p-values or effect sizes from completed analyses.
Export figures as SVG or PNG for manuscripts, slides, and reports.

Before exporting

Keep figures honest

Use clear axis labels that match your paper wording.
Choose grayscale when the figure may be printed without color.
Use summary tables for reporting; use charts for visual explanation.
Do not treat a formatted chart as a new analysis.

Figure CSV Upload data for a figure Use this when you want a standalone plotting file. You can also choose results already produced in other workspaces.

Choose figure CSV No file selected

Figure import settings

CSV delimiter

Use Auto for most files. Choose a delimiter manually if the preview shows one large column or shifted rows.

Figure examples

Download and upload one to try the builder

Grouped bar or boxplot figure_group_scores.csv has condition, group, score, reaction time, and accuracy columns.

Download

Repeated-measure line plot figure_repeated_measures.csv shows pre/post scores by condition for line charts.

Download

Scatter plot relation figure_scatter_relation.csv supports scatter plots for memory, reaction time, and accuracy.

Download

Data source

CSV or existing results

Choose where the figure should read data from. Existing results appear only after you run those analyses.

Source

Presentation guardrail

No new inferential claims

This workspace summarizes and formats selected data. Any p-values and effect sizes in result tables come from the original analysis workspace.

Upload a CSV or run an analysis in another workspace to begin.

Chart builder

Create a publication-ready chart

Chart type X variable Y variable Group variable Aggregation Error bars

Title X axis label Y axis label Color mode Font size Line / point size

Show legend

Figure preview

Exports match this SVG preview

Table builder

Create a clean results table

Use uploaded data for mean ± SD tables, or reuse p-values and effect sizes from completed toolkit analyses.

Table source Summary level Numeric variable Group variable

Table title Table theme Decimals Font size Row spacing Preview style

Show table title Alternate row shading

Table preview

Readable copy/export format

Local text comparison

Local Similarity Studio

Compare one target paper with a temporary pool of uploaded source texts. The app works locally against this session's files only and does not search the web.

No source pool loaded

What this module does

Source-limited similarity risk

Keeps uploaded source files ready for the current browser session.
Looks for long matching passages and shorter similar phrases.
Reports a similarity-risk score, source contributions, and the strongest matched passages.
Exports a match report for review or supervision notes.

Interpretation guardrail

Not proof of misconduct

A high percentage means text overlap with uploaded sources, not intent or final misconduct.
Translated paraphrases and semantic reuse are limited unless wording overlaps.
Documents are read locally when possible; scanned image-only files may need text recognition first.

Source pool Upload source texts Choose documents, text files, tables, or saved web pages. Uploading a new pool replaces the previous one in this session. Allowed formats: .pdf, .docx, .docm, .txt, .md, .markdown, .csv, .tsv, .html, .htm.

Choose source files No source files selected

Source pool limits

At capacity, the app asks users to retry later.

100 MBSession corpus 100Source files 25Web pages

Web page links

Paste or add source page links

0 / 25 links

Add links one by one. Internal and private links are blocked.

Source pool

Temporary session corpus

Method notes

Local text comparison

Upload source texts to begin.

Target paper Upload paper to check The target is compared with the current source pool and then summarized as a similarity-risk report. Allowed formats: .pdf, .docx, .docm, .txt, .md, .markdown, .csv, .tsv, .html, .htm.

Choose target text No target file selected

Analyze overlap

Review matched passages manually. The score combines long shared passages with shorter similar phrases from uploaded sources.

Source contributions

Matched target coverage

Strongest passages

Review before concluding

Choose a workspace

Stimulus Norming Studio

Participant Analysis Studio

Study Planning Studio

Results & Figures Studio

Local Similarity Studio

Upload a stimulus CSV

Clean stimulus file checklist

Included examples

CSV import settings

Preview

QC warnings

Map columns and confirm the dataset

Numeric metrics

Ignored columns

Type overrides

Run normalization analysis

Interpretation guardrails

Metric overview

Metric results

Row annotations

Generate counterbalanced lists and export files

List balance

Assignments

Participant Analysis Studio

What this module does

Participant CSV checklist

Try the example CSV

Suggested workflow

Import settings

Participant preview

QC warnings

Define participant data columns

Analysis mode

Outcome columns

Questionnaire item columns

Covariates

Repeated observations helper

Run participant checks

Row-wise structure

Repeated observations result

Group tests

Reliability

Exploratory factor suggestions

Item loading pattern

Scientific notes

Study Planning Studio

What this module supports

Planning checklist

Planning question

Planning curve

Assumptions and limits

Results & Figures Studio

What this module does

Before exporting

Figure import settings

Figure examples

Data source

Presentation guardrail

Create a publication-ready chart

Figure preview

Create a clean results table

Table preview

Local Similarity Studio

What this module does

Interpretation guardrail

Source pool limits

Web page links

Source pool

Method notes

Analyze overlap

Source contributions

Strongest passages