> [!warning] > Priyanka: The following Employee's moved campaigns - 7, 157, 161, 419, 529 and 605. They will appear twice in the Participant's Data sheet # day 2 - 432 completed day 2 completion day/time ![[s20220704_193817.png]] 26 subjects completed day 2 more than once. Use their first attempt (exclude subsequent attempts). ```r > d20time[, .N, id][N != 1] id N 1: 125 2 2: 142 3 3: 147 2 4: 195 2 5: 206 2 6: 207 2 7: 239 2 8: 245 3 9: 257 2 10: 277 2 11: 286 2 12: 311 2 13: 319 2 14: 329 2 15: 341 3 16: 357 2 17: 364 2 18: 369 2 19: 408 2 20: 459 2 21: 479 2 22: 484 2 23: 504 2 24: 517 2 25: 541 2 26: 572 2 id N ``` # day 1 - 413 completed day 1 ![[s20220704_194241.png]] 47 subjects completed day 1 more than once. - For those whose different attempts are the same survey, took their final attempt. - For those whose different attempts are different surveys, excluded them. ```r id N 1: 6 2 2: 125 2 3: 130 2 4: 181 5 5: 194 4 6: 199 2 7: 200 2 8: 201 5 9: 206 4 10: 209 2 11: 238 2 12: 240 2 13: 245 3 14: 257 2 15: 274 2 16: 301 2 17: 303 2 18: 355 2 19: 357 2 20: 385 2 21: 388 2 22: 408 2 23: 423 2 24: 433 2 25: 498 3 26: 550 3 27: 598 2 28: 195 2 29: 277 2 30: 302 2 31: 258 2 32: 288 2 33: 319 2 34: 286 3 35: 434 2 36: 519 2 37: 538 2 38: 38 2 39: 252 2 40: 289 2 41: 373 2 42: 404 2 43: 415 2 44: 492 2 45: 500 2 46: 501 2 47: 504 2 id N ``` - day 1 final n: 376 ```r # same no. of responses per subject > dt1[, .N, id][, table(N)] N 28 376 > dt1[, .N, keyby = .(id, counterbalance, condition)][, .N, keyby = .(condition)] condition N 1: accuracy 260 2: interest 116 > dt1[, .N, keyby = .(id, counterbalance, condition)][, .N, keyby = .(counterbalance, condition)] counterbalance condition N 1: 2 accuracy 106 2: 3 interest 29 3: 4 accuracy 24 4: 5 interest 29 5: 6 accuracy 106 6: 7 interest 21 7: 8 accuracy 24 8: 9 interest 37 ``` # how many completed both days - 301 completed both days - dropout: 75 completed day 1 but not day 2 - weird: 131 completed day 2 but not day 1 ``` # completed both days > bothdays <- intersect(day1id, day2id) > length(bothdays) [1] 301 # completed day 1 but not day 2 (dropout) > setdiff(day1id, day2id) [1] 29 30 39 104 105 107 109 112 123 124 149 166 169 180 184 188 196 203 221 229 230 243 252 254 259 264 271 272 275 280 284 295 310 313 314 [36] 333 339 340 346 353 356 358 363 374 381 382 395 396 401 404 435 436 444 445 454 462 477 493 500 501 503 512 523 524 527 532 543 559 566 573 [71] 577 583 584 594 599 > # dropout > setdiff(day1id, day2id) |> length() [1] 75 # weird ones: completed day 2 but not day 1 > setdiff(day2id, day1id) [1] 1 24 38 102 108 113 114 117 120 125 127 130 140 143 144 146 148 150 153 159 163 164 167 168 174 181 182 192 193 194 195 201 206 208 209 [36] 210 216 220 226 234 238 240 250 251 253 257 258 265 274 277 283 286 288 289 301 302 303 305 319 326 334 342 343 349 355 357 360 373 379 388 [71] 390 393 394 398 408 418 427 428 433 437 438 441 443 448 450 452 453 455 464 465 467 468 470 473 474 488 489 496 498 499 502 505 510 515 516 [106] 518 519 531 535 536 537 538 546 548 550 556 560 561 564 565 567 568 574 579 580 581 586 587 591 592 598 > # weird ones > setdiff(day2id, day1id) |> length() [1] 131 ``` ## dropout/attrition ```r > day1_dropout counterbalance condition N dropout_n dropout_prop 1: 2 accuracy 106 23 0.2169811 2: 3 interest 29 5 0.1724138 3: 4 accuracy 24 5 0.2083333 4: 5 interest 29 5 0.1724138 5: 6 accuracy 106 25 0.2358491 6: 7 interest 21 2 0.0952381 7: 8 accuracy 24 6 0.2500000 8: 9 interest 37 4 0.1081081 > prop.test(day1_dropout$dropout_n, day1_dropout$N) 8-sample test for equality of proportions without continuity correction data: day1_dropout$dropout_n out of day1_dropout$N X-squared = 5.1064, df = 7, p-value = 0.647 alternative hypothesis: two.sided sample estimates: prop 1 prop 2 prop 3 prop 4 prop 5 prop 6 prop 7 prop 8 0.2169811 0.1724138 0.2083333 0.1724138 0.2358491 0.0952381 0.2500000 0.1081081 > summary(aov(dropout_prop ~ counterbalance, day1_dropout)) Df Sum Sq Mean Sq F value Pr(>F) counterbalance 1 0.002515 0.002515 0.751 0.419 Residuals 6 0.020096 0.003349 # without grouping by counterbalance survey > prop.test(day1_dropout$dropout_n, day1_dropout$N) 2-sample test for equality of proportions with continuity correction data: day1_dropout$dropout_n out of day1_dropout$N X-squared = 3.4405, df = 1, p-value = 0.06362 alternative hypothesis: two.sided 95 percent confidence interval: 0.001952484 0.176031601 sample estimates: prop 1 prop 2 0.2269231 0.1379310 ``` # those who completed both days ```r > d5[!is.na(gap_days), .N, condition] condition N 1: accuracy 199 2: interest 99 ``` # gap between days ![[s20220704_210044.png]] exclude people who completed day 2 survey within 30 mins after completing day 1 survey ![[s20220704_211830.png]] # final subjects? - completed both days - completed day 2 at least 30 mins after completing day 1 ```r > d6[, .N, condition] condition N 1: accuracy 221 2: interest 107 ``` # descriptives ![[s20220704_213319.png]] ![[s20220704_213331.png]] ![[s20220704_213342.png]] ![[s20220704_213440.png]] experience (days) ![[s20220704_213506.png]]