This R markdown document provides R code and output for multiple datasets when using the CROSSTABS function in the Crosstabs.Loglinear package.
library(Crosstabs.Loglinear)**************************************************************************************************
Crosstabs.Loglinear 0.1.1
Please contact Brian O'Connor at brian.oconnor@ubc.ca if you have questions or suggestions.
**************************************************************************************************# when 'data' is a raw data file (rather than counts/frequencies)
CROSSTABS(data = subset(datasets$Field_2018_raw, Animal=='Cat'),
data_type = 'raw',
variables=c('Training','Dance') )
Observed Frequencies:
danced did not dance
affection 48 114
food 28 10
Row Totals:affection food
162 38
Column Totals: danced did not dance
76 124
N = 200
Expected Frequencies:
danced did not dance
affection 61.56 100.44
food 14.44 23.56
Number of cells with expected count < 5 = 0 (0%)
Chi-Square Tests: df p
Pearson Chi-Square 25.356 1 0
Yates Continuity Correction 23.520 1 0
Likelihood Ratio 24.932 1 0
Fisher's Exact p NA NA 0
Linear-by-Linear Assocn. 25.229 1 0
McNemar Test 50.880 1 0
Model Effect Sizes:
Contingency coefficient C 0.335
Phi 0.356
Cramer's V (from X2) 0.356
Cramer's V (from G2) 0.353
Cohen's W (from X2) 0.356
Cohen's W (from G2) 0.353
Residuals:
danced did not dance
affection -13.56 13.56
food 13.56 -13.56
Standardized Residuals:
danced did not dance
affection -1.728 1.353
food 3.568 -2.794
Adjusted Residuals:
danced did not dance
affection -5.035 5.035
food 5.035 -5.035
2-by-2 Table Effect Sizes$`Effect Sizes for affection vs. food () when = danced`
CI_lb CI_ub
Risk difference -0.441 -0.597 -0.284
Risk ratio 0.402 0.297 0.545
Odds ratio 0.150 0.068 0.334
Yule's Q -0.739 -0.873 -0.500
$`Effect Sizes for food vs. affection () when = danced`
CI_lb CI_ub
Risk difference 0.441 0.284 0.597
Risk ratio 2.487 1.835 3.370
Odds ratio 6.650 2.997 14.754
Yule's Q 0.739 0.500 0.873
$`Effect Sizes for affection vs. food () when = did not dance`
CI_lb CI_ub
Risk difference 0.441 0.284 0.597
Risk ratio 2.674 1.556 4.595
Odds ratio 6.650 2.997 14.754
Yule's Q 0.739 0.500 0.873
$`Effect Sizes for food vs. affection () when = did not dance`
CI_lb CI_ub
Risk difference -0.441 -0.597 -0.284
Risk ratio 0.374 0.218 0.643
Odds ratio 0.150 0.068 0.334
Yule's Q -0.739 -0.873 -0.500# when 'data' is a file with the counts/frequencies (rather than raw data points)
CROSSTABS(data = subset(datasets$Field_2018, Animal=='Cat'),
data_type = 'counts',
variables=c('Training','Dance') )
Observed Frequencies: Dance
Training danced did not dance
affection 48 114
food 28 10
Row Totals:affection food
162 38
Column Totals: danced did not dance
76 124
N = 200
Expected Frequencies: Dance
Training danced did not dance
affection 61.56 100.44
food 14.44 23.56
Number of cells with expected count < 5 = 0 (0%)
Chi-Square Tests: df p
Pearson Chi-Square 25.356 1 0
Yates Continuity Correction 23.520 1 0
Likelihood Ratio 24.932 1 0
Fisher's Exact p NA NA 0
Linear-by-Linear Assocn. 25.229 1 0
McNemar Test 50.880 1 0
Model Effect Sizes:
Contingency coefficient C 0.335
Phi 0.356
Cramer's V (from X2) 0.356
Cramer's V (from G2) 0.353
Cohen's W (from X2) 0.356
Cohen's W (from G2) 0.353
Residuals: Dance
Training danced did not dance
affection -13.56 13.56
food 13.56 -13.56
Standardized Residuals: Dance
Training danced did not dance
affection -1.728 1.353
food 3.568 -2.794
Adjusted Residuals: Dance
Training danced did not dance
affection -5.035 5.035
food 5.035 -5.035
2-by-2 Table Effect Sizes$`Effect Sizes for affection vs. food (Training) when Dance = danced`
CI_lb CI_ub
Risk difference -0.441 -0.597 -0.284
Risk ratio 0.402 0.297 0.545
Odds ratio 0.150 0.068 0.334
Yule's Q -0.739 -0.873 -0.500
$`Effect Sizes for food vs. affection (Training) when Dance = danced`
CI_lb CI_ub
Risk difference 0.441 0.284 0.597
Risk ratio 2.487 1.835 3.370
Odds ratio 6.650 2.997 14.754
Yule's Q 0.739 0.500 0.873
$`Effect Sizes for affection vs. food (Training) when Dance = did not dance`
CI_lb CI_ub
Risk difference 0.441 0.284 0.597
Risk ratio 2.674 1.556 4.595
Odds ratio 6.650 2.997 14.754
Yule's Q 0.739 0.500 0.873
$`Effect Sizes for food vs. affection (Training) when Dance = did not dance`
CI_lb CI_ub
Risk difference -0.441 -0.597 -0.284
Risk ratio 0.374 0.218 0.643
Odds ratio 0.150 0.068 0.334
Yule's Q -0.739 -0.873 -0.500# create and enter a two-dimensional contingency table for 'data'
food <- c(28, 10)
affection <- c(48, 114)
Field_2018_cats_conTable <- rbind(food, affection)
colnames(Field_2018_cats_conTable) <- c('danced', 'did not dance')
# add dimension names to the table
names(attributes(Field_2018_cats_conTable)$dimnames) <- c('Training','Dance')
CROSSTABS(data = Field_2018_cats_conTable, data_type = 'cont.table')
Observed Frequencies: Dance
Training danced did not dance
food 28 10
affection 48 114
Row Totals: food affection
38 162
Column Totals: danced did not dance
76 124
N = 200
Expected Frequencies: Dance
Training danced did not dance
food 14.44 23.56
affection 61.56 100.44
Number of cells with expected count < 5 = 0 (0%)
Chi-Square Tests: df p
Pearson Chi-Square 25.356 1 0
Yates Continuity Correction 23.520 1 0
Likelihood Ratio 24.932 1 0
Fisher's Exact p NA NA 0
Linear-by-Linear Assocn. 25.229 1 0
McNemar Test 23.603 1 0
Model Effect Sizes:
Contingency coefficient C 0.335
Phi 0.356
Cramer's V (from X2) 0.356
Cramer's V (from G2) 0.353
Cohen's W (from X2) 0.356
Cohen's W (from G2) 0.353
Residuals: Dance
Training danced did not dance
food 13.56 -13.56
affection -13.56 13.56
Standardized Residuals: Dance
Training danced did not dance
food 3.568 -2.794
affection -1.728 1.353
Adjusted Residuals: Dance
Training danced did not dance
food 5.035 -5.035
affection -5.035 5.035
2-by-2 Table Effect Sizes$`Effect Sizes for food vs. affection (Training) when Dance = danced`
CI_lb CI_ub
Risk difference 0.441 0.284 0.597
Risk ratio 2.487 1.835 3.370
Odds ratio 6.650 2.997 14.754
Yule's Q 0.739 0.500 0.873
$`Effect Sizes for affection vs. food (Training) when Dance = danced`
CI_lb CI_ub
Risk difference -0.441 -0.597 -0.284
Risk ratio 0.402 0.297 0.545
Odds ratio 0.150 0.068 0.334
Yule's Q -0.739 -0.873 -0.500
$`Effect Sizes for food vs. affection (Training) when Dance = did not dance`
CI_lb CI_ub
Risk difference -0.441 -0.597 -0.284
Risk ratio 0.374 0.218 0.643
Odds ratio 0.150 0.068 0.334
Yule's Q -0.739 -0.873 -0.500
$`Effect Sizes for affection vs. food (Training) when Dance = did not dance`
CI_lb CI_ub
Risk difference 0.441 0.284 0.597
Risk ratio 2.674 1.556 4.595
Odds ratio 6.650 2.997 14.754
Yule's Q 0.739 0.500 0.873# another way of creating the same two-dimensional contingency table for 'data'
Field_2018_cats_conTable_2 <- matrix( c(28, 48, 10, 114), nrow = 2, ncol = 2)
colnames(Field_2018_cats_conTable_2) <- c('danced', 'did not dance')
rownames(Field_2018_cats_conTable_2) <- c('food', 'affection')
CROSSTABS(data = Field_2018_cats_conTable_2, data_type = 'cont.table')
Observed Frequencies: danced did not dance
food 28 10
affection 48 114
Row Totals: food affection
38 162
Column Totals: danced did not dance
76 124
N = 200
Expected Frequencies: danced did not dance
food 14.44 23.56
affection 61.56 100.44
Number of cells with expected count < 5 = 0 (0%)
Chi-Square Tests: df p
Pearson Chi-Square 25.356 1 0
Yates Continuity Correction 23.520 1 0
Likelihood Ratio 24.932 1 0
Fisher's Exact p NA NA 0
Linear-by-Linear Assocn. 25.229 1 0
McNemar Test 23.603 1 0
Model Effect Sizes:
Contingency coefficient C 0.335
Phi 0.356
Cramer's V (from X2) 0.356
Cramer's V (from G2) 0.353
Cohen's W (from X2) 0.356
Cohen's W (from G2) 0.353
Residuals: danced did not dance
food 13.56 -13.56
affection -13.56 13.56
Standardized Residuals: danced did not dance
food 3.568 -2.794
affection -1.728 1.353
Adjusted Residuals: danced did not dance
food 5.035 -5.035
affection -5.035 5.035
2-by-2 Table Effect Sizes$`Effect Sizes for food vs. affection () when = danced`
CI_lb CI_ub
Risk difference 0.441 0.284 0.597
Risk ratio 2.487 1.835 3.370
Odds ratio 6.650 2.997 14.754
Yule's Q 0.739 0.500 0.873
$`Effect Sizes for affection vs. food () when = danced`
CI_lb CI_ub
Risk difference -0.441 -0.597 -0.284
Risk ratio 0.402 0.297 0.545
Odds ratio 0.150 0.068 0.334
Yule's Q -0.739 -0.873 -0.500
$`Effect Sizes for food vs. affection () when = did not dance`
CI_lb CI_ub
Risk difference -0.441 -0.597 -0.284
Risk ratio 0.374 0.218 0.643
Odds ratio 0.150 0.068 0.334
Yule's Q -0.739 -0.873 -0.500
$`Effect Sizes for affection vs. food () when = did not dance`
CI_lb CI_ub
Risk difference 0.441 0.284 0.597
Risk ratio 2.674 1.556 4.595
Odds ratio 6.650 2.997 14.754
Yule's Q 0.739 0.500 0.873CROSSTABS(data = subset(datasets$Field_2018, Animal=='Dog'),
data_type = 'counts',
variables=c('Training','Dance') )
Observed Frequencies: Dance
Training danced did not dance
affection 29 7
food 20 14
Row Totals:affection food
36 34
Column Totals: danced did not dance
49 21
N = 70
Expected Frequencies: Dance
Training danced did not dance
affection 25.2 10.8
food 23.8 10.2
Number of cells with expected count < 5 = 0 (0%)
Chi-Square Tests: df p
Pearson Chi-Square 3.932 1 0.04736
Yates Continuity Correction 2.966 1 0.08505
Likelihood Ratio 3.984 1 0.04594
Fisher's Exact p NA NA 0.06797
Linear-by-Linear Assocn. 3.876 1 0.04897
McNemar Test 5.333 1 0.02092
Model Effect Sizes:
Contingency coefficient C 0.231
Phi 0.237
Cramer's V (from X2) 0.237
Cramer's V (from G2) 0.239
Cohen's W (from X2) 0.237
Cohen's W (from G2) 0.239
Residuals: Dance
Training danced did not dance
affection 3.8 -3.8
food -3.8 3.8
Standardized Residuals: Dance
Training danced did not dance
affection 0.757 -1.156
food -0.779 1.190
Adjusted Residuals: Dance
Training danced did not dance
affection 1.983 -1.983
food -1.983 1.983
2-by-2 Table Effect Sizes$`Effect Sizes for affection vs. food (Training) when Dance = danced`
CI_lb CI_ub
Risk difference 0.217 0.007 0.427
Risk ratio 1.369 0.991 1.893
Odds ratio 2.900 0.993 8.466
Yule's Q 0.487 -0.003 0.789
$`Effect Sizes for food vs. affection (Training) when Dance = danced`
CI_lb CI_ub
Risk difference -0.217 -0.427 -0.007
Risk ratio 0.730 0.528 1.009
Odds ratio 0.345 0.118 1.007
Yule's Q -0.487 -0.789 0.003
$`Effect Sizes for affection vs. food (Training) when Dance = did not dance`
CI_lb CI_ub
Risk difference -0.217 -0.427 -0.007
Risk ratio 0.472 0.217 1.027
Odds ratio 0.345 0.118 1.007
Yule's Q -0.487 -0.789 0.003
$`Effect Sizes for food vs. affection (Training) when Dance = did not dance`
CI_lb CI_ub
Risk difference 0.217 0.007 0.427
Risk ratio 2.118 0.974 4.605
Odds ratio 2.900 0.993 8.466
Yule's Q 0.487 -0.003 0.789CROSSTABS(data = datasets$Howell_2017, data_type = 'counts', variables=c('Heart_Attack','Drug'))
Observed Frequencies: Drug
Heart_Attack Aspirin Placebo
No 10933 10845
Yes 104 189
Row Totals: No Yes
21778 293
Column Totals:Aspirin Placebo
11037 11034
N = 22071
Expected Frequencies: Drug
Heart_Attack Aspirin Placebo
No 10890.48 10887.52
Yes 146.52 146.48
Number of cells with expected count < 5 = 0 (0%)
Chi-Square Tests: df p
Pearson Chi-Square 25.014 1 0
Yates Continuity Correction 24.429 1 0
Likelihood Ratio 25.372 1 0
Fisher's Exact p NA NA 0
Linear-by-Linear Assocn. 25.013 1 0
McNemar Test 10534.989 1 0
Model Effect Sizes:
Contingency coefficient C 0.034
Phi 0.034
Cramer's V (from X2) 0.034
Cramer's V (from G2) 0.034
Cohen's W (from X2) 0.034
Cohen's W (from G2) 0.034
Residuals: Drug
Heart_Attack Aspirin Placebo
No 42.52 -42.52
Yes -42.52 42.52
Standardized Residuals: Drug
Heart_Attack Aspirin Placebo
No 0.407 -0.408
Yes -3.513 3.513
Adjusted Residuals: Drug
Heart_Attack Aspirin Placebo
No 5.001 -5.001
Yes -5.001 5.001
2-by-2 Table Effect Sizes$`Effect Sizes for No vs. Yes (Heart_Attack) when Drug = Aspirin`
CI_lb CI_ub
Risk difference 0.147 0.092 0.202
Risk ratio 1.414 1.211 1.651
Odds ratio 1.832 1.440 2.331
Yule's Q 0.294 0.180 0.400
$`Effect Sizes for Yes vs. No (Heart_Attack) when Drug = Aspirin`
CI_lb CI_ub
Risk difference -0.147 -0.202 -0.092
Risk ratio 0.707 0.606 0.826
Odds ratio 0.546 0.429 0.694
Yule's Q -0.294 -0.400 -0.180
$`Effect Sizes for No vs. Yes (Heart_Attack) when Drug = Placebo`
CI_lb CI_ub
Risk difference -0.147 -0.202 -0.092
Risk ratio 0.772 0.708 0.841
Odds ratio 0.546 0.429 0.694
Yule's Q -0.294 -0.400 -0.180
$`Effect Sizes for Yes vs. No (Heart_Attack) when Drug = Placebo`
CI_lb CI_ub
Risk difference 0.147 0.092 0.202
Risk ratio 1.295 1.189 1.412
Odds ratio 1.832 1.440 2.331
Yule's Q 0.294 0.180 0.400# change the order of the variables, which may or may not make the output more interpretable
CROSSTABS(data = datasets$Howell_2017, data_type = 'counts', variables=c('Drug','Heart_Attack'))
Observed Frequencies: Heart_Attack
Drug No Yes
Aspirin 10933 104
Placebo 10845 189
Row Totals:Aspirin Placebo
11037 11034
Column Totals: No Yes
21778 293
N = 22071
Expected Frequencies: Heart_Attack
Drug No Yes
Aspirin 10890.48 146.52
Placebo 10887.52 146.48
Number of cells with expected count < 5 = 0 (0%)
Chi-Square Tests: df p
Pearson Chi-Square 25.014 1 0
Yates Continuity Correction 24.429 1 0
Likelihood Ratio 25.372 1 0
Fisher's Exact p NA NA 0
Linear-by-Linear Assocn. 25.013 1 0
McNemar Test 10534.989 1 0
Model Effect Sizes:
Contingency coefficient C 0.034
Phi 0.034
Cramer's V (from X2) 0.034
Cramer's V (from G2) 0.034
Cohen's W (from X2) 0.034
Cohen's W (from G2) 0.034
Residuals: Heart_Attack
Drug No Yes
Aspirin 42.52 -42.52
Placebo -42.52 42.52
Standardized Residuals: Heart_Attack
Drug No Yes
Aspirin 0.407 -3.513
Placebo -0.408 3.513
Adjusted Residuals: Heart_Attack
Drug No Yes
Aspirin 5.001 -5.001
Placebo -5.001 5.001
2-by-2 Table Effect Sizes$`Effect Sizes for Aspirin vs. Placebo (Drug) when Heart_Attack = No`
CI_lb CI_ub
Risk difference 0.008 0.005 0.011
Risk ratio 1.008 1.005 1.011
Odds ratio 1.832 1.440 2.331
Yule's Q 0.294 0.180 0.400
$`Effect Sizes for Placebo vs. Aspirin (Drug) when Heart_Attack = No`
CI_lb CI_ub
Risk difference -0.008 -0.011 -0.005
Risk ratio 0.992 0.989 0.995
Odds ratio 0.546 0.429 0.694
Yule's Q -0.294 -0.400 -0.180
$`Effect Sizes for Aspirin vs. Placebo (Drug) when Heart_Attack = Yes`
CI_lb CI_ub
Risk difference -0.008 -0.011 -0.005
Risk ratio 0.550 0.434 0.698
Odds ratio 0.546 0.429 0.694
Yule's Q -0.294 -0.400 -0.180
$`Effect Sizes for Placebo vs. Aspirin (Drug) when Heart_Attack = Yes`
CI_lb CI_ub
Risk difference 0.008 0.005 0.011
Risk ratio 1.818 1.433 2.306
Odds ratio 1.832 1.440 2.331
Yule's Q 0.294 0.180 0.400CROSSTABS(data = datasets$Noursis_2012_marital,
data_type = 'counts',
variables=c('Marital_Status','Gen.Happiness'))
Observed Frequencies: Gen.Happiness
Marital_Status Happy Not Happy
Married 566 38
Split 320 72
Never Married 313 60
Row Totals: Married Split Never Married
604 392 373
Column Totals: Happy Not Happy
1199 170
N = 1369
Expected Frequencies: Gen.Happiness
Marital_Status Happy Not Happy
Married 528.996 75.004
Split 343.322 48.678
Never Married 326.682 46.318
Number of cells with expected count < 5 = 0 (0%)
Chi-Square Tests: df p
Pearson Chi-Square 38.217 2 0
Yates Continuity Correction 38.217 2 0
Likelihood Ratio 40.480 2 0
Fisher's Exact p NA NA 0
Linear-by-Linear Association 25.163 2 0
Model Effect Sizes:
Contingency coefficient C 0.165
Phi 0.167
Cramer's V (from X2) 0.167
Cramer's V (from G2) 0.172
Cohen's W (from X2) 0.167
Cohen's W (from G2) 0.172
Residuals: Gen.Happiness
Marital_Status Happy Not Happy
Married 37.004 -37.004
Split -23.322 23.322
Never Married -13.682 13.682
Standardized Residuals: Gen.Happiness
Marital_Status Happy Not Happy
Married 1.609 -4.273
Split -1.259 3.343
Never Married -0.757 2.010
Adjusted Residuals: Gen.Happiness
Marital_Status Happy Not Happy
Married 6.108 -6.108
Split -4.228 4.228
Never Married -2.518 2.518CROSSTABS(data = datasets$Noursis_2012_voting_degree,
data_type = 'counts',
variables=c('Vote','College.Degree'))
Observed Frequencies: College.Degree
Vote No Yes
No 369 50
Yes 659 372
Row Totals: No Yes
419 1031
Column Totals: No Yes
1028 422
N = 1450
Expected Frequencies: College.Degree
Vote No Yes
No 297.057 121.943
Yes 730.943 300.057
Number of cells with expected count < 5 = 0 (0%)
Chi-Square Tests: df p
Pearson Chi-Square 84.199 1 0
Yates Continuity Correction 83.033 1 0
Likelihood Ratio 94.241 1 0
Fisher's Exact p NA NA 0
Linear-by-Linear Assocn. 84.141 1 0
McNemar Test 521.388 1 0
Model Effect Sizes:
Contingency coefficient C 0.234
Phi 0.241
Cramer's V (from X2) 0.241
Cramer's V (from G2) 0.255
Cohen's W (from X2) 0.241
Cohen's W (from G2) 0.255
Residuals: College.Degree
Vote No Yes
No 71.943 -71.943
Yes -71.943 71.943
Standardized Residuals: College.Degree
Vote No Yes
No 4.174 -6.515
Yes -2.661 4.153
Adjusted Residuals: College.Degree
Vote No Yes
No 9.176 -9.176
Yes -9.176 9.176
2-by-2 Table Effect Sizes$`Effect Sizes for No vs. Yes (Vote) when College.Degree = No`
CI_lb CI_ub
Risk difference 0.241 0.199 0.284
Risk ratio 1.378 1.300 1.460
Odds ratio 4.166 3.020 5.746
Yule's Q 0.613 0.503 0.704
$`Effect Sizes for Yes vs. No (Vote) when College.Degree = No`
CI_lb CI_ub
Risk difference -0.241 -0.284 -0.199
Risk ratio 0.726 0.685 0.769
Odds ratio 0.240 0.174 0.331
Yule's Q -0.613 -0.704 -0.503
$`Effect Sizes for No vs. Yes (Vote) when College.Degree = Yes`
CI_lb CI_ub
Risk difference -0.241 -0.284 -0.199
Risk ratio 0.331 0.252 0.434
Odds ratio 0.240 0.174 0.331
Yule's Q -0.613 -0.704 -0.503
$`Effect Sizes for Yes vs. No (Vote) when College.Degree = Yes`
CI_lb CI_ub
Risk difference 0.241 0.199 0.284
Risk ratio 3.024 2.302 3.971
Odds ratio 4.166 3.020 5.746
Yule's Q 0.613 0.503 0.704CROSSTABS(data = datasets$Stevens_2009_HeadStart_1, data_type = 'counts',
variables=c('SEX', 'ATTITUDE'))
Observed Frequencies: ATTITUDE
SEX 1 2
1 33 7
2 37 23
Row Totals: 1 2
40 60
Column Totals: 1 2
70 30
N = 100
Expected Frequencies: ATTITUDE
SEX 1 2
1 28 12
2 42 18
Number of cells with expected count < 5 = 0 (0%)
Chi-Square Tests: df p
Pearson Chi-Square 4.960 1 0.02594
Yates Continuity Correction 4.018 1 0.04502
Likelihood Ratio 5.194 1 0.02266
Fisher's Exact p NA NA 0.02833
Linear-by-Linear Assocn. 4.911 1 0.02669
McNemar Test 19.114 1 0.00001
Model Effect Sizes:
Contingency coefficient C 0.217
Phi 0.223
Cramer's V (from X2) 0.223
Cramer's V (from G2) 0.228
Cohen's W (from X2) 0.223
Cohen's W (from G2) 0.228
Residuals: ATTITUDE
SEX 1 2
1 5 -5
2 -5 5
Standardized Residuals: ATTITUDE
SEX 1 2
1 0.945 -1.443
2 -0.772 1.179
Adjusted Residuals: ATTITUDE
SEX 1 2
1 2.227 -2.227
2 -2.227 2.227
2-by-2 Table Effect Sizes$`Effect Sizes for 1 vs. 2 (SEX) when ATTITUDE = 1`
CI_lb CI_ub
Risk difference 0.208 0.038 0.379
Risk ratio 1.338 1.047 1.710
Odds ratio 2.931 1.114 7.711
Yule's Q 0.491 0.054 0.770
$`Effect Sizes for 2 vs. 1 (SEX) when ATTITUDE = 1`
CI_lb CI_ub
Risk difference -0.208 -0.379 -0.038
Risk ratio 0.747 0.585 0.955
Odds ratio 0.341 0.130 0.898
Yule's Q -0.491 -0.770 -0.054
$`Effect Sizes for 1 vs. 2 (SEX) when ATTITUDE = 2`
CI_lb CI_ub
Risk difference -0.208 -0.379 -0.038
Risk ratio 0.457 0.217 0.962
Odds ratio 0.341 0.130 0.898
Yule's Q -0.491 -0.770 -0.054
$`Effect Sizes for 2 vs. 1 (SEX) when ATTITUDE = 2`
CI_lb CI_ub
Risk difference 0.208 0.038 0.379
Risk ratio 2.190 1.039 4.616
Odds ratio 2.931 1.114 7.711
Yule's Q 0.491 0.054 0.770CROSSTABS(data = datasets$Warner_2020_titanic,
data_type = 'cont.table')
Observed Frequencies: Died Survived
First 4 140
Second 13 80
Third 89 76
Row Totals: First Second Third
144 93 165
Column Totals: Died Survived
106 296
N = 402
Expected Frequencies: Died Survived
First 37.970 106.030
Second 24.522 68.478
Third 43.507 121.493
Number of cells with expected count < 5 = 0 (0%)
Chi-Square Tests: df p
Pearson Chi-Square 113.231 2 0
Yates Continuity Correction 113.231 2 0
Likelihood Ratio 124.286 2 0
Fisher's Exact p NA NA 0
Linear-by-Linear Association 105.362 2 0
Model Effect Sizes:
Contingency coefficient C 0.469
Phi 0.531
Cramer's V (from X2) 0.531
Cramer's V (from G2) 0.556
Cohen's W (from X2) 0.531
Cohen's W (from G2) 0.556
Residuals: Died Survived
First -33.970 33.970
Second -11.522 11.522
Third 45.493 -45.493
Standardized Residuals: Died Survived
First -5.513 3.299
Second -2.327 1.392
Third 6.897 -4.127
Adjusted Residuals: Died Survived
First -8.019 8.019
Second -3.093 3.093
Third 10.468 -10.468CROSSTABS(data = datasets$Warner_2020_dog,
data_type = 'cont.table')
Observed Frequencies: Died Survived
No 11 28
Yes 3 50
Row Totals: No Yes
39 53
Column Totals: Died Survived
14 78
N = 92
Expected Frequencies: Died Survived
No 5.935 33.065
Yes 8.065 44.935
Number of cells with expected count < 5 = 0 (0%)
Chi-Square Tests: df p
Pearson Chi-Square 8.851 1 0.00293
Yates Continuity Correction 7.190 1 0.00733
Likelihood Ratio 9.011 1 0.00268
Fisher's Exact p NA NA 0.00637
Linear-by-Linear Assocn. 8.755 1 0.00309
McNemar Test 18.581 1 0.00002
Model Effect Sizes:
Contingency coefficient C 0.296
Phi 0.310
Cramer's V (from X2) 0.310
Cramer's V (from G2) 0.313
Cohen's W (from X2) 0.310
Cohen's W (from G2) 0.313
Residuals: Died Survived
No 5.065 -5.065
Yes -5.065 5.065
Standardized Residuals: Died Survived
No 2.079 -0.881
Yes -1.784 0.756
Adjusted Residuals: Died Survived
No 2.975 -2.975
Yes -2.975 2.975
2-by-2 Table Effect Sizes$`Effect Sizes for No vs. Yes () when = Died`
CI_lb CI_ub
Risk difference 0.225 0.071 0.380
Risk ratio 4.983 1.489 16.673
Odds ratio 6.548 1.684 25.456
Yule's Q 0.735 0.255 0.924
$`Effect Sizes for Yes vs. No () when = Died`
CI_lb CI_ub
Risk difference -0.225 -0.380 -0.071
Risk ratio 0.201 0.060 0.672
Odds ratio 0.153 0.039 0.594
Yule's Q -0.735 -0.924 -0.255
$`Effect Sizes for No vs. Yes () when = Survived`
CI_lb CI_ub
Risk difference -0.225 -0.380 -0.071
Risk ratio 0.761 0.618 0.936
Odds ratio 0.153 0.039 0.594
Yule's Q -0.735 -0.924 -0.255
$`Effect Sizes for Yes vs. No () when = Survived`
CI_lb CI_ub
Risk difference 0.225 0.071 0.380
Risk ratio 1.314 1.068 1.617
Odds ratio 6.548 1.684 25.456
Yule's Q 0.735 0.255 0.924