forked from AdaLovelaceInstitute/how-do-people-feel-about-ai
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathPCA_Regression_How do people feel about AI.R
661 lines (638 loc) · 28.6 KB
/
PCA_Regression_How do people feel about AI.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
#
###################################################################################################
# Attitudes to AI Survey 2023
# Full R script for combined regression plot
###################################################################################################
#
####################################################################################################
# 1. General data management to create initial 'analysis_data' dataset
####################################################################################################
#
remove(list = ls())
setwd("..")
install.packages("questionr")
install.packages("gmodels")
install.packages("psych")
install.packages("GPArotation")
install.packages("summarytools")
install.packages("foreign")
install.packages("fmsb")
install.packages("labelled")
install.packages("dotwhisker")
install.packages("tidyverse")
library(questionr)
library(gmodels)
library(psych)
library(GPArotation)
library(summarytools)
library(foreign)
library(fmsb)
library(labelled)
library(dotwhisker)
library(tidyverse)
#
# Read in original data files from Kantar
ai_raw_data <- read_csv("262400371_Public Attitudes to AI Raw Data NA.csv")
#
# rename items to more interpretable names
ai_raw_data <- rename(ai_raw_data, ben_face_unlock = BENA_MP
, con_face_unlock = CONA_MP
, ben_face_border = BENA_BC
, con_face_border = CONA_BC
, ben_face_police = BENA_PS
, con_face_police = CONA_PS
, ben_welfare = BENA_WEL
, con_welfare = CONA_WEL
, ben_job = BENA_JOB
, con_job = CONA_JOB
, ben_cancer = BENA_CAN
, con_cancer = CONA_CAN
, ben_loan = BENA_LOAN
, con_loan = CONA_LOAN
, ben_socmedia_consum = BENA_SMC
, con_socmedia_consum = CONA_SMC
, ben_socmedia_politad = BENA_SMP
, con_socmedia_politad = CONA_SMP
, ben_virt_assist_speaker = BENA_VASS
, con_virt_assist_speaker = CONA_VASS
, ben_virt_assist_health = BENA_VAH
, con_virt_assist_health = CONA_VAH
, ben_rob_vacuum = BENA_RVC
, con_rob_vacuum = CONA_RVC
, ben_rob_care_assist = BENA_RCA
, con_rob_care_assist = CONA_RCA
, ben_driverless_cars = BENA_DC
, con_driverless_cars = CONA_DC
, ben_auto_weapons = BENA_AW
, con_auto_weapons = CONA_AW
, ben_climate_res = BENA_CCR
, con_climate_res = CONA_CCR
, ben_educ = BENA_VR
, con_educ = CONA_VR
)
#
group1 <- c("ben_face_unlock", "con_face_unlock", "ben_face_police", "con_face_police", "ben_job", "con_job", "ben_loan", "con_loan", "ben_socmedia_consum", "con_socmedia_consum", "ben_virt_assist_speaker", "con_virt_assist_speaker", "ben_rob_vacuum", "con_rob_vacuum", "ben_driverless_cars", "con_driverless_cars", "ben_climate_res", "con_climate_res")
group2 <- c("ben_face_unlock", "con_face_unlock", "ben_face_border", "con_face_border", "ben_welfare", "con_welfare", "ben_cancer", "con_cancer", "ben_socmedia_politad", "con_socmedia_politad", "ben_virt_assist_health", "con_virt_assist_health", "ben_rob_care_assist", "con_rob_care_assist", "ben_auto_weapons", "con_auto_weapons", "ben_educ", "con_educ")
#
# recode codes 5 and 6 (DK and Ref) as missing
ai_raw_data <- ai_raw_data %>%
mutate_at(group1, funs(na_if(.,6)))
ai_raw_data <- ai_raw_data %>%
mutate_at(group1, funs(na_if(.,5)))
ai_raw_data <- ai_raw_data %>%
mutate_at(group2, funs(na_if(.,6)))
ai_raw_data <- ai_raw_data %>%
mutate_at(group2, funs(na_if(.,5)))
#
keep.these <- c(
"Respondent_ID",
"PV25_Weight",
"CS_Age",
"CS_Age_Band",
"RS_Sex",
"RS_EducationLevel",
"RS_EmploymentStatus_MostRecent",
"RS_NSSEC_SummaryCat",
"DR_Socio_Eco",
"RS_InternetUseFrequency",
"RS_NumTechDevices",
"TECHINTE",
"TECHINFO",
"DIG_LIT_1",
"DIG_LIT_2",
"DIG_LIT_3",
"DIG_LIT_4",
"DIG_LIT_5",
"DIG_LIT_6",
"DIG_LIT_7",
"DIG_LIT_8",
"TECHSELF_1",
"TECHSELF_2",
"TECHSELF_3",
"TECHSELF_4",
"TECHSELF_5",
"TECHSOCIAL_1",
"TECHSOCIAL_2",
"TECHINTE_NET",
"TECHINFO_NET",
"DR_DIGLIT_CONFIDENCE",
"DR_TECHCOMFORT",
"RS_RespectUKValues",
"RS_BanInternetSites",
"RS_ProtectWorkingConditionsWages",
"RS_BusinessOverregulated",
"RS_NationalisePublicServicesIndustries",
"RS_HumanRightProtest",
"RS_TolerateUnconventional",
"RS_TougherSentences",
"RS_RedistributeIncome",
"RS_BigBusinessBenefitsOwners",
"RS_FewerImmigrants",
"RS_AbolishMonarchy",
"RS_BFI_2_XS_ExtraversionScore",
"RS_BFI_2_XS_AgreeablenessScore",
"RS_BFI_2_XS_ConscientiousnessScore",
"RS_BFI_2_XS_NegativeEmotionalityScore",
"RS_BFI_2_XS_OpenMindednessScore",
"AWARE_MP",
"AWARE_BC",
"AWARE_PS",
"AWARE_WEL",
"AWARE_JOB",
"AWARE_CAN",
"AWARE_LOAN",
"AWARE_SMC",
"AWARE_SMP",
"AWARE_VASS",
"AWARE_VAH",
"AWARE_RVC",
"AWARE_RCA",
"AWARE_DC",
"AWARE_AW",
"AWARE_CCR",
"AWARE_VR",
"EXP_MP",
"EXP_BC",
"EXP_PS",
"EXP_WEL",
"EXP_JOB",
"EXP_CAN",
"EXP_LOAN",
"EXP_SMC",
"EXP_SMP",
"EXP_VASS",
"EXP_VAH",
"EXP_RVC",
"EXP_VR",
"ben_face_unlock",
"con_face_unlock",
"ben_face_police",
"con_face_police",
"ben_job",
"con_job",
"ben_loan",
"con_loan",
"ben_virt_assist_speaker",
"con_virt_assist_speaker",
"ben_rob_vacuum",
"con_rob_vacuum",
"ben_driverless_cars",
"con_driverless_cars",
"ben_climate_res",
"con_climate_res",
"ben_face_border",
"con_face_border",
"ben_welfare",
"con_welfare",
"ben_cancer",
"con_cancer",
"ben_socmedia_politad",
"con_socmedia_politad",
"ben_virt_assist_health",
"con_virt_assist_health",
"ben_rob_care_assist",
"con_rob_care_assist",
"ben_auto_weapons",
"con_auto_weapons",
"ben_educ",
"con_educ",
"ben_socmedia_consum",
"con_socmedia_consum"
)
ai_raw_data <- ai_raw_data[keep.these]
#
# Saving the file
saveRDS(ai_raw_data, "analysis_data.rds")
#
#################################################################################
# 2. Demographic variables recoding
#################################################################################
#
analysis_data <- readRDS("analysis_data.rds")
#
library(questionr)
library(tidyverse)
library(dplyr)
library(gmodels)
#
# delete 16yrd old and Sex=Other
analysis_data<-analysis_data[!(analysis_data$CS_Age==16 | analysis_data$RS_Sex==3),]
#
# age band dummies
analysis_data$Age_band_factor <- factor(analysis_data$CS_Age_Band)
library(forcats)
# Combine age bands
analysis_data %>%
mutate(age_band_new = fct_collapse(Age_band_factor,
"16-34" = c("1", "2"),
"35-64" = c("3", "4", "5"),
"65 and above" = c("6", "7")
)) -> analysis_data
#
# dummies for age
analysis_data$age65_and_over <- ifelse(analysis_data$age_band_new=="65 and above",1,0)
freq(analysis_data$age65_and_over, cum = TRUE, total = TRUE)
#
# recode RS_Sex into dummy
analysis_data$Male <- analysis_data$RS_Sex
analysis_data %>%
mutate(Male=ifelse(Male==1,1,0),
) -> analysis_data
#
# Techinfo and Techinte recoding
analysis_data$TECHINFO[analysis_data$TECHINFO==5] <- NA
analysis_data$TECHINFO[analysis_data$TECHINFO==6] <- NA
analysis_data$TECHINTE[analysis_data$TECHINTE==5] <- NA
analysis_data$TECHINTE[analysis_data$TECHINTE==6] <- NA
#
# reverse code
temp_vars <- c("TECHINFO","TECHINTE")
analysis_data[ , temp_vars] = 5 - analysis_data[ , temp_vars]
#
# Education variable recoding
analysis_data %>%
mutate(Degree=ifelse(RS_EducationLevel==1,1,0),
) -> analysis_data
#
# analysis_data1 includes all key variables recoded
saveRDS(analysis_data, "analysis_data1.rds")
#
#################################################################################
# 3. Principle Component Analysis
#################################################################################
#
analysis_data1 <- readRDS("analysis_data1.rds")
# PCA1: Digital literacy
itemsdig <- c("DIG_LIT_1", "DIG_LIT_2", "DIG_LIT_3", "DIG_LIT_4", "DIG_LIT_5", "DIG_LIT_6", "DIG_LIT_7", "DIG_LIT_8")
# Don't know and Prefer not to say recoded into missing
analysis_data1 <- analysis_data1 %>%
mutate_at(itemsdig, funs(na_if(.,5)))
analysis_data1 <- analysis_data1 %>%
mutate_at(itemsdig, funs(na_if(.,6)))
apply((analysis_data1[c(itemsdig)]), 2, table)
#
# reverse code
analysis_data1[ , itemsdig] = 5 - analysis_data1[ , itemsdig]
apply((analysis_data1[c(itemsdig)]), 2, table)
# correlation
round(cor(analysis_data1[,itemsdig],use="complete.obs"),4)
# PCA
PCA_DigLit <- princomp(~DIG_LIT_1+DIG_LIT_2+DIG_LIT_3+DIG_LIT_4+DIG_LIT_5+DIG_LIT_6+DIG_LIT_7+DIG_LIT_8,
data=analysis_data1, cor = TRUE, scores = TRUE, na.action=na.exclude)
summary(PCA_DigLit)
PCA_DigLit$sdev^2
PCA_DigLit.weights <- loadings(PCA_DigLit)
print(PCA_DigLit.weights,cutoff=0,digits=4)
sqrt.lambda <- PCA_DigLit$sdev
#
# Loadings
print(t(t(PCA_DigLit.weights)*sqrt.lambda),cutoff = 0,digits=4)
#
# Responses and PC scores 1 and 2 for first 10 respondents
cbind(analysis_data1[,itemsdig], round(PCA_DigLit$score[,1:2],3))[1:10,]
#
analysis_data1 <- cbind(analysis_data1, PCA_DigLit$score)
analysis_data1 <- analysis_data1 %>%
rename(digital_comp1 = Comp.1)
summary(analysis_data1$digital_comp1)
#
drop <- c("Comp.2","Comp.3","Comp.4","Comp.5","Comp.6","Comp.7","Comp.8")
analysis_data1 = analysis_data1[,!(names(analysis_data1) %in% drop)]
#
# PCA2: Comfortable with technology (7 items)
itemscomf <- c("TECHSELF_1", "TECHSELF_2", "TECHSELF_3", "TECHSELF_4", "TECHSELF_5", "TECHSOCIAL_1", "TECHSOCIAL_2")
# correlation
round(cor(analysis_data1[,itemscomf],use="complete.obs"),4)
# PCA
PCA_TechComf <- princomp(~TECHSELF_1+TECHSELF_2+TECHSELF_3+TECHSELF_4+TECHSELF_5+TECHSOCIAL_1+TECHSOCIAL_2,
data=analysis_data1, cor = TRUE, scores = TRUE, na.action=na.exclude)
summary(PCA_TechComf)
PCA_TechComf$sdev^2
PCA_TechComf.weights <- loadings(PCA_TechComf)
print(PCA_TechComf.weights,cutoff=0,digits=4)
#
sqrt.lambda <- PCA_TechComf$sdev
# Loadings
print(t(t(PCA_TechComf.weights)*sqrt.lambda),cutoff = 0,digits=4)
#
# Responses and PC scores 1 and 2 for first 10 respondents
cbind(analysis_data1[,itemscomf], round(PCA_TechComf$score[,1:2],3))[1:10,]
#
analysis_data1 <- cbind(analysis_data1, PCA_TechComf$score)
analysis_data1 <- analysis_data1 %>%
rename(techcomf7_comp1 = Comp.1)
summary(analysis_data1$techcomf7_comp1)
#
drop <- c("Comp.2","Comp.3","Comp.4","Comp.5","Comp.6","Comp.7")
analysis_data1 = analysis_data1[,!(names(analysis_data1) %in% drop)]
names(analysis_data1)
# analysis_data2 includes all key variables recoded and PCA Comp1 scores for diglit and techcomf
saveRDS(analysis_data1, "analysis_data2.rds")
#
#################################################################################
# 4. Further recoding for combined regression coefficients plot
#################################################################################
#
# Open data file for regression analysis
analysis_data2 <- readRDS("analysis_data2.rds")
#
# create a new variable 'net_benefit'=benefit - concern
net_vars <- c("ben_face_unlock","con_face_unlock","ben_face_police","con_face_police","ben_job","con_job","ben_loan"
,"con_loan","ben_virt_assist_speaker","con_virt_assist_speaker","ben_rob_vacuum","con_rob_vacuum"
,"ben_driverless_cars","con_driverless_cars","ben_climate_res","con_climate_res","ben_face_border"
,"con_face_border","ben_welfare","con_welfare","ben_cancer","con_cancer","ben_socmedia_politad"
,"con_socmedia_politad","ben_virt_assist_health","con_virt_assist_health","ben_rob_care_assist"
,"con_rob_care_assist","ben_auto_weapons","con_auto_weapons","ben_educ","con_educ","ben_socmedia_consum"
,"con_socmedia_consum")
# reverse code all benefit and concern variables
analysis_data2[ , net_vars] = 5 - analysis_data2[ , net_vars]
#
mutate(analysis_data2,
net_ben_face_unlock = ben_face_unlock - con_face_unlock,
net_ben_face_police = ben_face_police - con_face_police,
net_ben_job = ben_job - con_job,
net_ben_loan = ben_loan - con_loan,
net_ben_virt_assist_speaker = ben_virt_assist_speaker - con_virt_assist_speaker,
net_ben_rob_vacuum = ben_rob_vacuum - con_rob_vacuum,
net_ben_driverless_cars = ben_driverless_cars - con_driverless_cars,
net_ben_climate_res = ben_climate_res - con_climate_res,
net_ben_face_border = ben_face_border - con_face_border,
net_ben_welfare = ben_welfare - con_welfare,
net_ben_cancer = ben_cancer - con_cancer,
net_ben_socmedia_politad = ben_socmedia_politad - con_socmedia_politad,
net_ben_virt_assist_health = ben_virt_assist_health - con_virt_assist_health,
net_ben_rob_care_assist = ben_rob_care_assist - con_rob_care_assist,
net_ben_auto_weapons = ben_auto_weapons - con_auto_weapons,
net_ben_educ = ben_educ - con_educ,
net_ben_socmedia_consum = ben_socmedia_consum - con_socmedia_consum
) -> analysis_data2
#
new_net_vars <- c("net_ben_face_unlock","net_ben_face_police","net_ben_job","net_ben_loan"
,"net_ben_virt_assist_speaker","net_ben_rob_vacuum","net_ben_driverless_cars","net_ben_climate_res","net_ben_face_border"
,"net_ben_welfare","net_ben_cancer","net_ben_socmedia_politad","net_ben_virt_assist_health","net_ben_rob_care_assist"
,"net_ben_auto_weapons","net_ben_educ","net_ben_socmedia_consum")
#
# analysis_data3 includes net_benefit for all 17 techs (previous version analysis_data2 includes variables recoded and PCA Comp1 scores for diglit, techcomf
saveRDS(analysis_data2, "analysis_data3.rds")
#
# Open cleaned data file for regression analysis
analysis_data3 <- readRDS("analysis_data3.rds")
#
# Recoding variable Aware of individual technology to a dummy
#
analysis_data3$AWARE_MP_dummy <- analysis_data3$AWARE_MP
analysis_data3$AWARE_BC_dummy <- analysis_data3$AWARE_BC
analysis_data3$AWARE_PS_dummy <- analysis_data3$AWARE_PS
analysis_data3$AWARE_WEL_dummy <- analysis_data3$AWARE_WEL
analysis_data3$AWARE_JOB_dummy <- analysis_data3$AWARE_JOB
analysis_data3$AWARE_CAN_dummy <- analysis_data3$AWARE_CAN
analysis_data3$AWARE_LOAN_dummy <- analysis_data3$AWARE_LOAN
analysis_data3$AWARE_SMC_dummy <- analysis_data3$AWARE_SMC
analysis_data3$AWARE_SMP_dummy <- analysis_data3$AWARE_SMP
analysis_data3$AWARE_VASS_dummy <- analysis_data3$AWARE_VASS
analysis_data3$AWARE_VAH_dummy <- analysis_data3$AWARE_VAH
analysis_data3$AWARE_RVC_dummy <- analysis_data3$AWARE_RVC
analysis_data3$AWARE_RCA_dummy <- analysis_data3$AWARE_RCA
analysis_data3$AWARE_DC_dummy <- analysis_data3$AWARE_DC
analysis_data3$AWARE_AW_dummy <- analysis_data3$AWARE_AW
analysis_data3$AWARE_CCR_dummy <- analysis_data3$AWARE_CCR
analysis_data3$AWARE_VR_dummy <- analysis_data3$AWARE_VR
#
# vector with variables
aware_vars_dummy <- c("AWARE_MP_dummy","AWARE_BC_dummy","AWARE_PS_dummy","AWARE_WEL_dummy","AWARE_JOB_dummy","AWARE_CAN_dummy","AWARE_LOAN_dummy","AWARE_SMC_dummy"
,"AWARE_SMP_dummy","AWARE_VASS_dummy","AWARE_VAH_dummy","AWARE_RVC_dummy","AWARE_RCA_dummy","AWARE_DC_dummy","AWARE_AW_dummy","AWARE_CCR_dummy","AWARE_VR_dummy")
#
# create a function ro recoding into dummy all Aware variables
dummy_funct <- function(x) {
case_when(
{{x}} == 1 ~ 1,
{{x}} == 2 | {{x}} == 3 ~ 0,
TRUE ~ NA
)
}
# use dummy_function across all variables
analysis_data3 %>%
mutate(across(all_of(aware_vars_dummy), ~ dummy_funct(.x))) -> analysis_data3
#
# Recoding of Experience of technology variables
analysis_data3$EXP_MP_dummy <- analysis_data3$EXP_MP
analysis_data3$EXP_BC_dummy <- analysis_data3$EXP_BC
analysis_data3$EXP_PS_dummy <- analysis_data3$EXP_PS
analysis_data3$EXP_WEL_dummy <- analysis_data3$EXP_WEL
analysis_data3$EXP_JOB_dummy <- analysis_data3$EXP_JOB
analysis_data3$EXP_CAN_dummy <- analysis_data3$EXP_CAN
analysis_data3$EXP_LOAN_dummy <- analysis_data3$EXP_LOAN
analysis_data3$EXP_SMC_dummy <- analysis_data3$EXP_SMC
analysis_data3$EXP_SMP_dummy <- analysis_data3$EXP_SMP
analysis_data3$EXP_VASS_dummy <- analysis_data3$EXP_VASS
analysis_data3$EXP_VAH_dummy <- analysis_data3$EXP_VAH
analysis_data3$EXP_RVC_dummy <- analysis_data3$EXP_RVC
analysis_data3$EXP_VR_dummy <- analysis_data3$EXP_VR
#
exp_vars_dummy <- c("EXP_MP_dummy","EXP_BC_dummy","EXP_PS_dummy","EXP_WEL_dummy","EXP_JOB_dummy","EXP_CAN_dummy","EXP_LOAN_dummy","EXP_SMC_dummy"
,"EXP_SMP_dummy","EXP_VASS_dummy","EXP_VAH_dummy","EXP_RVC_dummy","EXP_VR_dummy")
#
# create a function
dummy_funct1 <- function(x) {
case_when(
{{x}} == 1 | {{x}} == 2 ~ 1,
{{x}} == 3 | {{x}} == 4 ~ 0,
TRUE ~ NA
)
}
# use dummy_function across all variables
analysis_data3 %>%
mutate(across(all_of(exp_vars_dummy), ~ dummy_funct1(.x))) -> analysis_data3
#
# Social class recoding RS_NSSEC_SummaryCat
analysis_data3$RS_NSSEC_SummaryCat_factor <- factor(analysis_data3$RS_NSSEC_SummaryCat)
#
# Combine age bands
library(forcats)
freq(analysis_data3$RS_NSSEC_SummaryCat_factor, cum = TRUE, total = TRUE)
analysis_data3 %>%
mutate(RS_NSSEC_SummaryCat_new = fct_collapse(RS_NSSEC_SummaryCat_factor,
"NSSEC_1-3" = c("1", "2"),
"NSSEC_4-5" = c("3", "4", "5"),
"NSSEC_6-7" = c("6", "7")
)) -> analysis_data3
#
# dummy for NC SEC
analysis_data3$NSSEC_1_3 <- ifelse(analysis_data3$RS_NSSEC_SummaryCat_new=="NSSEC_1-3",1,0)
#
# save as analysis_data4
saveRDS(analysis_data3, "analysis_data4.rds")
#
#################################################################################
# 5. Combined regression plot
#################################################################################
# open analysis_data4 to run the code for the combined plot
#
analysis_data4 <- readRDS("analysis_data4.rds")
#
# new 17 dummy variables on whether respondent was asked about a technology
analysis_data4$AWARE_MP_asked <- analysis_data4$AWARE_MP
analysis_data4$AWARE_BC_asked <- analysis_data4$AWARE_BC
analysis_data4$AWARE_PS_asked <- analysis_data4$AWARE_PS
analysis_data4$AWARE_WEL_asked <- analysis_data4$AWARE_WEL
analysis_data4$AWARE_JOB_asked <- analysis_data4$AWARE_JOB
analysis_data4$AWARE_CAN_asked <- analysis_data4$AWARE_CAN
analysis_data4$AWARE_LOAN_asked <- analysis_data4$AWARE_LOAN
analysis_data4$AWARE_SMC_asked <- analysis_data4$AWARE_SMC
analysis_data4$AWARE_SMP_asked <- analysis_data4$AWARE_SMP
analysis_data4$AWARE_VASS_asked <- analysis_data4$AWARE_VASS
analysis_data4$AWARE_VAH_asked <- analysis_data4$AWARE_VAH
analysis_data4$AWARE_RVC_asked <- analysis_data4$AWARE_RVC
analysis_data4$AWARE_RCA_asked <- analysis_data4$AWARE_RCA
analysis_data4$AWARE_DC_asked <- analysis_data4$AWARE_DC
analysis_data4$AWARE_AW_asked <- analysis_data4$AWARE_AW
analysis_data4$AWARE_CCR_asked <- analysis_data4$AWARE_CCR
analysis_data4$AWARE_VR_asked <- analysis_data4$AWARE_VR
#
# vector with variables
aware_vars_asked <- c("AWARE_MP_asked","AWARE_BC_asked","AWARE_PS_asked","AWARE_WEL_asked","AWARE_JOB_asked","AWARE_CAN_asked","AWARE_LOAN_asked"
,"AWARE_SMC_asked","AWARE_SMP_asked","AWARE_VASS_asked","AWARE_VAH_asked","AWARE_RVC_asked","AWARE_RCA_asked","AWARE_DC_asked"
,"AWARE_AW_asked","AWARE_CCR_asked","AWARE_VR_asked")
#
# create a function for a new dummy variable (Aware Q asked or not)
dummy_funct2 <- function(x) {
case_when(
{{x}} == 1 | {{x}} == 2 | {{x}} == 3 | {{x}} == 4 ~ 1,
TRUE ~ 0
)
}
# use dummy_function across all variables
analysis_data4 %>%
mutate(across(all_of(aware_vars_asked), ~ dummy_funct2(.x))) -> analysis_data4
#
# regression for each technology,
# each regression uses a subset on the sample that was asked about the technology.
subset_1 <- subset(analysis_data4, AWARE_VAH_asked == 1)
subset_1$AWARE <- subset_1$AWARE_VAH_dummy
subset_1$EXP <- subset_1$EXP_VAH_dummy
m1 <- lm(net_ben_virt_assist_health ~ age65_and_over + Male + Degree + AWARE + EXP + TECHINFO + TECHINTE + digital_comp1 + techcomf7_comp1 + NSSEC_1_3
, data = subset_1, weights=PV25_Weight)
subset_2 <- subset(analysis_data4, AWARE_VASS_asked == 1)
subset_2$AWARE <- subset_2$AWARE_VASS_dummy
subset_2$EXP <- subset_2$EXP_VASS_dummy
m2 <- lm(net_ben_virt_assist_speaker ~ age65_and_over + Male + Degree + AWARE + EXP + TECHINFO + TECHINTE + digital_comp1 + techcomf7_comp1 + NSSEC_1_3
, data = subset_2, weights=PV25_Weight)
subset_3 <- subset(analysis_data4, AWARE_SMC_asked == 1)
subset_3$AWARE <- subset_3$AWARE_SMC_dummy
subset_3$EXP <- subset_3$EXP_SMC_dummy
m3 <- lm(net_ben_socmedia_consum ~ age65_and_over + Male + Degree + AWARE + EXP + TECHINFO + TECHINTE + digital_comp1 + techcomf7_comp1 + NSSEC_1_3
, data = subset_3, weights=PV25_Weight)
subset_4 <- subset(analysis_data4, AWARE_SMP_asked == 1)
subset_4$AWARE <- subset_4$AWARE_SMP_dummy
subset_4$EXP <- subset_4$EXP_SMP_dummy
m4 <- lm(net_ben_socmedia_politad ~ age65_and_over + Male + Degree + AWARE + EXP + TECHINFO + TECHINTE + digital_comp1 + techcomf7_comp1 + NSSEC_1_3
, data = subset_4, weights=PV25_Weight)
subset_5 <- subset(analysis_data4, AWARE_VR_asked == 1)
subset_5$AWARE <- subset_5$AWARE_VR_dummy
subset_5$EXP <- subset_5$EXP_VR_dummy
m5 <- lm(net_ben_educ ~ age65_and_over + Male + Degree + AWARE + EXP + TECHINFO + TECHINTE + digital_comp1 + techcomf7_comp1 + NSSEC_1_3
, data = subset_5, weights=PV25_Weight)
subset_6 <- subset(analysis_data4, AWARE_CCR_asked == 1)
subset_6$AWARE <- subset_6$AWARE_CCR_dummy
m6 <- lm(net_ben_climate_res ~ age65_and_over + Male + Degree + AWARE + TECHINFO + TECHINTE + digital_comp1 + techcomf7_comp1 + NSSEC_1_3
, data = subset_6, weights=PV25_Weight)
subset_7 <- subset(analysis_data4, AWARE_RVC_asked == 1)
subset_7$AWARE <- subset_7$AWARE_RVC_dummy
subset_7$EXP <- subset_7$EXP_RVC_dummy
m7 <- lm(net_ben_rob_vacuum ~ age65_and_over + Male + Degree + AWARE + EXP + TECHINFO + TECHINTE + digital_comp1 + techcomf7_comp1 + NSSEC_1_3
, data = subset_7, weights=PV25_Weight)
subset_8 <- subset(analysis_data4, AWARE_RCA_asked == 1)
subset_8$AWARE <- subset_8$AWARE_RCA_dummy
m8 <- lm(net_ben_rob_care_assist ~ age65_and_over + Male + Degree + AWARE + TECHINFO + TECHINTE + digital_comp1 + techcomf7_comp1 + NSSEC_1_3
, data = subset_8, weights=PV25_Weight)
subset_9 <- subset(analysis_data4, AWARE_CAN_asked == 1)
subset_9$AWARE <- subset_9$AWARE_CAN_dummy
subset_9$EXP <- subset_9$EXP_CAN_dummy
m9 <- lm(net_ben_cancer ~ age65_and_over + Male + Degree + AWARE + EXP + TECHINFO + TECHINTE + digital_comp1 + techcomf7_comp1 + NSSEC_1_3
, data = subset_9, weights=PV25_Weight)
subset_10 <- subset(analysis_data4, AWARE_LOAN_asked == 1)
subset_10$AWARE <- subset_10$AWARE_LOAN_dummy
subset_10$EXP <- subset_10$EXP_LOAN_dummy
m10 <- lm(net_ben_loan ~ age65_and_over + Male + Degree + AWARE + EXP + TECHINFO + TECHINTE + digital_comp1 + techcomf7_comp1 + NSSEC_1_3
, data = subset_10, weights=PV25_Weight)
subset_11 <- subset(analysis_data4, AWARE_MP_asked == 1)
subset_11$AWARE <- subset_11$AWARE_MP_dummy
subset_11$EXP <- subset_11$EXP_MP_dummy
m11 <- lm(net_ben_face_unlock ~ age65_and_over + Male + Degree + AWARE + EXP + TECHINFO + TECHINTE + digital_comp1 + techcomf7_comp1 + NSSEC_1_3
, data = subset_11, weights=PV25_Weight)
subset_12 <- subset(analysis_data4, AWARE_PS_asked == 1)
subset_12$AWARE <- subset_12$AWARE_PS_dummy
subset_12$EXP <- subset_12$EXP_PS_dummy
m12 <- lm(net_ben_face_police ~ age65_and_over + Male + Degree + AWARE + EXP + TECHINFO + TECHINTE + digital_comp1 + techcomf7_comp1 + NSSEC_1_3
, data = subset_12, weights=PV25_Weight)
subset_13 <- subset(analysis_data4, AWARE_BC_asked == 1)
subset_13$AWARE <- subset_13$AWARE_BC_dummy
subset_13$EXP <- subset_13$EXP_BC_dummy
m13 <- lm(net_ben_face_border ~ age65_and_over + Male + Degree + AWARE + EXP + TECHINFO + TECHINTE + digital_comp1 + techcomf7_comp1 + NSSEC_1_3
, data = subset_13, weights=PV25_Weight)
subset_14 <- subset(analysis_data4, AWARE_WEL_asked == 1)
subset_14$AWARE <- subset_14$AWARE_WEL_dummy
subset_14$EXP <- subset_14$EXP_WEL_dummy
m14 <- lm(net_ben_welfare ~ age65_and_over + Male + Degree + AWARE + EXP + TECHINFO + TECHINTE + digital_comp1 + techcomf7_comp1 + NSSEC_1_3
, data = subset_14, weights=PV25_Weight)
subset_15 <- subset(analysis_data4, AWARE_JOB_asked == 1)
subset_15$AWARE <- subset_15$AWARE_JOB_dummy
subset_15$EXP <- subset_15$EXP_JOB_dummy
m15 <- lm(net_ben_job ~ age65_and_over + Male + Degree + AWARE + EXP + TECHINFO + TECHINTE + digital_comp1 + techcomf7_comp1 + NSSEC_1_3
, data = subset_15, weights=PV25_Weight)
subset_16 <- subset(analysis_data4, AWARE_AW_asked == 1)
subset_16$AWARE <- subset_16$AWARE_AW_dummy
m16 <- lm(net_ben_auto_weapons ~ age65_and_over + Male + Degree + AWARE + TECHINFO + TECHINTE + digital_comp1 + techcomf7_comp1 + NSSEC_1_3
, data = subset_16, weights=PV25_Weight)
subset_17 <- subset(analysis_data4, AWARE_DC_asked == 1)
subset_17$AWARE <- subset_17$AWARE_DC_dummy
m17 <- lm(net_ben_driverless_cars ~ age65_and_over + Male + Degree + AWARE + TECHINFO + TECHINTE + digital_comp1 + techcomf7_comp1 + NSSEC_1_3
, data = subset_17, weights=PV25_Weight)
#
# switch off scientific notations
options(scipen=999)
obj <- dwplot(list(m1,m2,m3,m4,m5,m6,m7,m8,m9,m10,m11,m12,m13,m14,m15,m16,m17))
obj
obj2 <- as.data.frame(obj$data)
obj2
obj2$model <- fct_recode(obj2$model, "Virtual assistant: health" = "Model 1", "Virtual assistant: speaker" = "Model 2", "Social media ads: product" = "Model 3"
,"Social media ads: political" = "Model 4","Simulation: VR education" = "Model 5", "Simulation: climate change" = "Model 6"
,"Robotics: vacuum" = "Model 7", "Robotics: care assistant" = "Model 8","Predict: risk cancer" = "Model 9"
,"Predict: loan" = "Model 10", "Face recognition: unlock" = "Model 11", "Face recognition: policing" = "Model 12"
,"Face recognition: border" = "Model 13","Eligibility: welfare" = "Model 14", "Eligibility: jobs" = "Model 15"
,"Autonomous: weapons" = "Model 16", "Autonomous: cars" = "Model 17")
#
unique(obj2$model)
#
# install.packages("scales")
library(scales)
#install.packages("ggpmisc")
library(ggpmisc)
ggplot(data = obj2, aes(x = estimate, y = model)) +
geom_point(color=ifelse(obj2$p.value<0.05, 'red', 'black')) +
labs(caption = "*Statistically significant coefficients (p<0.05) are marked in red\n**The question on experience was not asked for all technologies") +
theme(plot.caption = element_text(hjust=0)) +
geom_vline(xintercept=0,linetype=3.5) +
xlim(limits = c(-0.8, 0.8)) +
facet_wrap(~ term, nrow = 2, labeller = as_labeller(c(age65_and_over = "Aged 65+",
Male = "Male",
Degree = "Degree",
AWARE = "Awareness of technology",
EXP = "Experience with technology**",
TECHINFO = "Tech informed",
TECHINTE= "Tech interested",
digital_comp1= "Digital literacy",
techcomf7_comp1= "Tech comfortable",
NSSEC_1_3 = "Socio-economic class 1-3"))) +
scale_x_continuous(labels = label_number(accuracy = 0.1),
) +
xlab("Coefficient Estimates (higher values = more net benefit)") + ylab("AI Technologies") +
ggtitle("Predictors* of net benefit") +
theme(
plot.title = element_text(face = "bold"),
legend.position = c(0.82, 0.00),
legend.justification = c(0, 0),
legend.background = element_rect(colour = "grey80"),
legend.title = element_blank()
)
#