Only 6 studies have condition 1: B C G H L P
Included only studies with both conditions 0 and 1. We won't be looking at conditions > 1.
Preregistration: The total sample size is 6961 (control: 2867; experimental: 4094). Prior to this preregistration, we randomly sampled about 20% of participants from each condition within each study (6 in total) to perform exploratory analyses; thus, this preregistration focuses on the remaining 80% of participants (5633 participants: control: 2321; experimental: 3312).
After excluding outlier trials, we have **5338** participants in **fold 1**. Excluded 5.24% subjects (excluded n = 295 out of 5633) in **fold 1**.
- 5.51% in condition 0 (128 in fold 1)
- 5.04% in condition 1 (167 in fold 1)
```r
# cleaned data
fold condition N
1: 0 0 546
2: 0 1 782
3: 1 0 2193 #
4: 1 1 3145 #
# raw data
fold condition N
1: 0 0 546
2: 0 1 782
3: 1 0 2321
4: 1 1 3312
```
Raw data
```r
> dt2[condition %in% c(0, 1) & study %in% studies, n_distinct(id), keyby = .(fold, study, studyOrig, condition)]
fold study studyOrig condition V1
1: 0 B Pennycook Nature S3 0 114
2: 0 B Pennycook Nature S3 1 114
3: 0 C Pennycook Nature S4 0 122
4: 0 C Pennycook Nature S4 1 124
5: 0 G Pennycook Nature S5 0 118
6: 0 G Pennycook Nature S5 1 59
7: 0 H Reset S4 0 72
8: 0 H Reset S4 1 363
9: 0 L Epstein HKS 0 73
10: 0 L Epstein HKS 1 73
11: 0 P Arechar Pilot 0 47
12: 0 P Arechar Pilot 1 49
13: 1 B Pennycook Nature S3 0 463
14: 1 B Pennycook Nature S3 1 466
15: 1 C Pennycook Nature S4 0 494
16: 1 C Pennycook Nature S4 1 501
17: 1 G Pennycook Nature S5 0 527
18: 1 G Pennycook Nature S5 1 266
19: 1 H Reset S4 0 310
20: 1 H Reset S4 1 1534
21: 1 L Epstein HKS 0 314
22: 1 L Epstein HKS 1 322
23: 1 P Arechar Pilot 0 213
24: 1 P Arechar Pilot 1 223
fold study studyOrig condition V1
```
Cleaned data
```r
# cleaned fold 0
> d0[condition %in% c(0, 1), n_distinct(id), keyby = .(fold, condition, study, studyOrig)]
fold condition study studyOrig V1
1: 0 0 B Pennycook Nature S3 114
2: 0 0 C Pennycook Nature S4 122
3: 0 0 D Reset S1 37
4: 0 0 E Reset S2 39
5: 0 0 F Reset S3 33
6: 0 0 G Pennycook Nature S5 118
7: 0 0 H Reset S4 72
8: 0 0 L Epstein HKS 73
9: 0 0 P Arechar Pilot 47
10: 0 1 B Pennycook Nature S3 114
11: 0 1 C Pennycook Nature S4 124
12: 0 1 G Pennycook Nature S5 59
13: 0 1 H Reset S4 363
14: 0 1 L Epstein HKS 73
15: 0 1 P Arechar Pilot 49
# cleaned fold 1
> d1[condition %in% c(0, 1), n_distinct(id), keyby = .(fold, condition, study, studyOrig)]
fold condition study studyOrig V1
1: 1 0 B Pennycook Nature S3 459
2: 1 0 C Pennycook Nature S4 488
3: 1 0 D Reset S1 152
4: 1 0 E Reset S2 156
5: 1 0 F Reset S3 136
6: 1 0 G Pennycook Nature S5 472
7: 1 0 H Reset S4 290
8: 1 0 L Epstein HKS 296
9: 1 0 P Arechar Pilot 188
10: 1 1 B Pennycook Nature S3 461
11: 1 1 C Pennycook Nature S4 495
12: 1 1 G Pennycook Nature S5 240
13: 1 1 H Reset S4 1453
14: 1 1 L Epstein HKS 296
15: 1 1 P Arechar Pilot 200
```
Dave's raw data: `meta data w rts for dmm.csv`
```r
> dt0[, .N, by = .(studyOrig, id, study)][, .N, by = .(studyOrig, study)]
studyOrig study N
1: Arechar Pilot P 820
2: Epstein HKS L 1192
3: Pennycook Nature S3 B 1157
4: Pennycook Nature S4 C 1242
5: Pennycook Nature S5 G 1286
6: Reset S1 D 1007
7: Reset S2 E 1210
8: Reset S3 F 1184
9: Reset S4 H 2296
```