Mixed methods — civic participation, from grounded theory to a survey
Exploratory sequential design · 240 synthetic respondents (categories → items → α → EFA → regression)
Research question — Do the conditions of participation built in grounded theory appear with the same structure once moved into a survey, and how much do they explain intention to keep participating?
Raw data sample_mixed_participation.csv · 240 rows
Synthetic data — 240 responses generated by scripts/sample1/gen-mixed.mjs (seed 20260730). The statistics are measured values produced by scripts/sample1/compute_mixed.py (mixedComputed.generated.ts).
| id | gender | age | years_resident | eff1 | eff2 | eff3 | eff4 | own1 | own2 | own3 | bar1 | bar2 | bar3 | int1 | int2 | int3 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| S001 | F | 26 | 6 | 2 | 3 | 2 | 4 | 3 | 2 | 2 | 5 | 5 | 5 | 1 | 2 | 2 |
| S002 | F | 55 | 23 | 3 | 2 | 2 | 2 | 1 | 2 | 2 | 4 | 3 | 4 | 2 | 3 | 2 |
| S003 | F | 67 | 15 | 3 | 3 | 2 | 4 | 2 | 3 | 3 | 4 | 3 | 4 | 3 | 2 | 3 |
| S004 | M | 34 | 0 | 2 | 2 | 2 | 2 | 2 | 2 | 3 | 5 | 5 | 5 | 1 | 2 | 1 |
| S005 | M | 27 | 3 | 3 | 4 | 3 | 4 | 1 | 2 | 1 | 4 | 3 | 3 | 2 | 2 | 2 |
The study, step by step — what was actually produced
① Where it starts — the grounded-theory categories
This is an exploratory sequential design — the qualitative phase comes first, the quantitative phase after it. It starts from the categories built by the showcase case “Grounded theory — conditions for civic participation”: “trying to speak up and taking a small role”, “the tension between my own work and my share”, “the threshold set by how meetings run”, “how the notice reaches people”, “the air the existing members carry”, “the fork between staying, watching and leaving”.
② Categories → survey items
Each category was carried over into 5-point Likert items — participation efficacy 4 items, psychological ownership 3, barriers to participation 3, intention to keep participating 3 (the outcome), with age, years resident and gender as covariates.
“Trying to speak up and taking a small role” → eff1 “I can say what I think at our neighborhood meetings” · “The tension between my own work and my share” → own1 “What happens in this neighborhood feels like my business” · “The threshold set by how meetings run” → bar1 “The time of day meetings are held makes it hard for me to take part”
③ Collecting the synthetic responses
The 240 responses are synthetic — scripts/sample1/gen-mixed.mjs (fixed seed 20260730) generated them with a factor structure built in, and every statistic after that is a measured value produced by this product’s own engines running on the same data (the reliability, factor-analysis and regression engines themselves).
④ Reliability — Cronbach’s alpha
Alpha came out at 0.841 for participation efficacy, 0.849 for psychological ownership, 0.807 for barriers to participation and 0.846 for intention to keep participating.
⑤ Exploratory factor analysis (EFA)
Running the 10 items of efficacy, ownership and barriers as three factors (principal-component extraction, varimax rotation) separated the items of each scale onto its own factor — KMO 0.827, variance explained per factor 27.0% · 21.8% · 23.3%. The number of factors and the rotation stay yours to change in the run options.
⑥ Regression — intention to keep participating
Intention to keep participating was regressed on the three scale means plus age and years resident — participation efficacy b=0.421 (p < .001) · ownership b=0.335 (p < .001) · barriers b=-0.321 (p < .001) · age b=0.004 (p 0.319) · years resident b=0.008 (p 0.310), R²=0.589. The direction the qualitative phase drew — efficacy and ownership push, barriers pull — shows up in the signs of these coefficients as it was.
Key results
Categories → items
Where the categories of the qualitative phase became scales in the quantitative one.
| Grounded-theory category | Scale (example item) | Items |
|---|---|---|
| Trying to speak up and taking a small role | Participation efficacy (eff1 “I can say what I think at our neighborhood meetings”) | 4 |
| The tension between my own work and my share | Psychological ownership (own1 “What happens in this neighborhood feels like my business”) | 3 |
| The threshold set by how meetings run · how the notice reaches people · the air the existing members carry | Barriers to participation (bar1 “The time of day meetings are held makes it hard for me to take part”) | 3 |
| The fork between staying, watching and leaving | Intention to keep participating (int1 “I mean to keep going to neighborhood meetings”) — outcome | 3 |
Reliability (alpha)
Measured by compute_mixed.py — three decimal places.
| Scale | Items | Cronbach’s alpha |
|---|---|---|
| Participation efficacy | 4 | 0.841 |
| Psychological ownership | 3 | 0.849 |
| Barriers to participation | 3 | 0.807 |
| Intention to keep participating | 3 | 0.846 |
EFA loadings (three factors)
Principal-component extraction · varimax rotation · KMO 0.827 · variance explained per factor 27.0% · 21.8% · 23.3%.
| Item | Factor 1 | Factor 2 | Factor 3 |
|---|---|---|---|
| eff1 | 0.779 | -0.103 | 0.304 |
| eff2 | 0.781 | -0.151 | 0.163 |
| eff3 | 0.813 | -0.092 | 0.113 |
| eff4 | 0.793 | -0.066 | 0.220 |
| own1 | 0.240 | -0.077 | 0.844 |
| own2 | 0.249 | -0.066 | 0.831 |
| own3 | 0.167 | -0.116 | 0.853 |
| bar1 | -0.040 | 0.826 | -0.086 |
| bar2 | -0.152 | 0.852 | -0.051 |
| bar3 | -0.132 | 0.839 | -0.100 |
Regression — intention to keep participating
OLS · N=240 · R²=0.589.
| Variable | b | SE | p |
|---|---|---|---|
| (intercept) | 1.469 | 0.279 | < .001 |
| Participation efficacy (mean) | 0.421 | 0.053 | < .001 |
| Psychological ownership (mean) | 0.335 | 0.050 | < .001 |
| Barriers to participation (mean) | -0.321 | 0.050 | < .001 |
| Age | 0.004 | 0.004 | 0.319 |
| Years resident | 0.008 | 0.008 | 0.310 |
Representative quotes
Every quote is a sentence that exists in the raw data (excerpt-existence check passed).
Paper-ready tables
Copying opens an editable duplicate in your workbench — this showcase stays as it is.