Background
Main steps of the methodology of influence analysis concerning machine learning
|
---|
Step 1. Gathering questionnaire answers from persons representing various health and demographic backgrounds. - Each person gives the “need for help” ratings for a set of common expression statements that describe imagined scenarios. - The rating answers given by the person form his/her “need for help” rating profile. (Described in the chapter “Gathering ratings about expression statements from persons representing various background features”) |
Step 2. Identifying statistically significant and non-significant rating differences for expression statements in respect to groupings based on the answer values of background questions (for example groupings relying on the person’s answer about his/her estimated health condition). (Described in the chapter “Identifying statistically significant rating differences for expression statements in respect to background questions”) |
Step 3. Training and validation of a machine learning model (with a supervised learning approach) to learn the groupings concerning the “need for help” ratings. This step uses the same groupings of respondents that have been used in the step 2. (Described in the chapter “Training and validation of a machine learning model to learn groupings concerning the ratings”) |
Step 4. Comparing the validation accuracies of the machine learning model with the probabilities of pure chance of classifying the rating profiles correctly (averaged from at least 100 separate training and validation sequences). (Described in the chapter “Comparing the validation accuracies of the machine learning model with the probabilities of pure chance”) |
Step 5. Contrasting the validation accuracies of the machine learning model with the occurrence of statistically significant and non-significant rating differences for expression statements in respect to groupings based on the answer values of background questions (averaged from at least 100 separate training and validation sequences). (Described in the chapter “Contrasting the validation accuracies of the machine learning model with the statistically significant rating differences in respect to groupings”) |
Step 6. Drawing conclusions about the applicability of the current machine learning model in this knowledge context. Based on the conclusions further fitting of the model and iteratively repeating the steps 2-6. (Described in the chapter “Drawing conclusions about the applicability of the current machine learning model”) |
-
RQ1) How do different people rate the “need for help” for a set of health-related expression statements and how this rating depends on the background information about the person (such as his/her demographic information and evaluation about own health and wellbeing)? This main research question RQ1 emphasizes especially the steps 1-2 of Table 1.
-
RQ2) What kinds of results can be gained when training a convolutional neural network model based on the “need for help” ratings to classify persons into groups based on their background information? This main research question RQ2 emphasizes especially the steps 3-6 of Table 1.
Methods
1. Gathering ratings about expression statements from persons representing various background features
1.1 Study design, setting, participants and sampling strategy
1.2 Variables and study size
1.3 Data sources/measurement
Compact notation
|
Expression statement
|
Range of values for the person’s answer (indicating the “need for help” rating)
|
---|---|---|
ES1 | “I have a flu.” | 0-10 |
ES2 | “I have a cough.” | 0-10 |
ES3 | “I have a shortness of breath.” | 0-10 |
ES4 | “My health condition is weakening.” | 0-10 |
ES5 | “I have a sore throat.” | 0-10 |
ES6 | “I have muscular ache.” | 0-10 |
ES7 | “I have a fever.” | 0-10 |
ES8 | “A sudden fever rises for me with 38 degrees of Celsius or more.” | 0-10 |
ES9 | “I suspect that I have now become infected by the coronavirus.” | 0-10 |
ES10 | “I have now become infected by the coronavirus.” | 0-10 |
ES11 | “I am quarantined from meeting other people ordinarily so that the spreading of an infectious disease could be prevented.” | 0-10 |
ES12 | “I must be inside a house without getting out.” | 0-10 |
ES13 | “I must be without a human companion.” | 0-10 |
ES14 | “I do not cope in everyday life independently without getting help from other persons.” | 0-10 |
ES15 | “I do not cope at home independently without getting help from persons who originate outside of my home.” | 0-10 |
ES16 | “I have an infectious disease.” | 0-10 |
ES17 | “I have an infectious disease that has been verified by a doctor.” | 0-10 |
ES18 | “I suspect that I have an infectious disease.” | 0-10 |
ES19 | “I have a bad health condition.” | 0-10 |
ES20 | “I have an ordinary health condition.” | 0-10 |
Compact notation
|
Question about the person’s background information
|
Range of values for the person’s answer
|
---|---|---|
BQ1: an estimated health condition | A 9-point Likert scale supplied with the following partial labeling: “9 Good”, “8 –“, “7 Rather good”, “6 –“, “5 Medium”, “4 –“, “3 Rather bad”, “2 –“, “1 Bad”. | |
BQ2: a health problem reduces ability | “Do you have a permanent or long-lasting disease or such deficit, ailment or disability that reduces your ability to work or to perform your daily living activities? Here the question refers to all long-lasting diseases identified by a doctor, and also to such ailments not identified by a doctor which have lasted at least three months but which affect your ability to perform your daily living activities.“ [32] | No or yes. |
BQ3: one or more diseases identified by a doctor | “Has there been a situation that a doctor has identified in you one or several of the following diseases?“ [32] | The person answers by selecting one or more options from a list of diseases [18], see details in Data analysis supplement (Additional file 1). For some options there is a question “other, what?” and an adjacent text input box so that the person can write some additional information concerning that option. |
BQ4: a continuous or repeated need for a doctor’s care | “Do you need continuously or repeatedly care given by a doctor for a long-lasting disease, deficit or disability that you have just mentioned?“ [32] | No or yes. |
BQ5: the quality of life | A 9-point Likert scale supplied with the following partial labeling: “9 Very good”, “8 –“, “7 Good”, “6 –“, “5 Neither good nor bad”, “4 –“, “3 Bad”, “2 –“, “1 Very bad”. | |
BQ6: the satisfaction about health | A 9-point Likert scale supplied with the following partial labeling: “9 Very satisfied”, “8 –“, “7 Satisfied”, “6 –“, “5 Neither satisfied nor dissatisfied”, “4 –“, “3 Dissatisfied”, “2 –“, “1 Very dissatisfied”. | |
BQ7: the satisfaction about ability | A 9-point Likert scale supplied with the following partial labeling: “9 Very satisfied”, “8 –“, “7 Satisfied”, “6 –“, “5 Neither satisfied nor dissatisfied”, “4 –“, “3 Dissatisfied”, “2 –“, “1 Very dissatisfied”. | |
BQ8: the sex | “Tell what is your sex. The answer alternatives are similar as in the earlier health surveys of Finnish Institute for Health and Welfare (THL) to maintain comparability with the earlier results.” [32] | Man or woman. |
BQ9: the age | “Tell what is your age.” [32] | Age in years selected from the following range: 16 years, 17 years, …, 99 years, 100 years or more. |
1.4 Bias
1.5 Quantitative variables and statistical methods
Statistically significant rating differences
|
Training and validation metrics
2
of the convolutional neural network model to learn a labeling that matches the grouping
|
Comparison of the validation accuracy with the probability of pure chance
| ||||||
---|---|---|---|---|---|---|---|---|
Grouping based on the answer value (x) of the background question (BQ)
|
Expression statements (ES) having statistically significant rating differences in the grouping (the difference of mean ratings
1
about the need for help)
|
Epoch step
|
Training loss
|
Training accuracy
|
Validation loss
|
Validation accuracy
|
Probability of pure chance of classifying the rating profiles correctly
3
(based on the size of the greatest group)
|
Difference of the mean validation accuracy and the probability of pure chance of classifying the rating profiles correctly
|
BQ1, two groups: x < 7 (n1=263), x>=7 (n2=410) | ES6: diffg1&g2=0,07 (95% CI [0,03; 0,11], p = 0.0011); ES8: diffg1&g2=-0,08 (95% CI [-0,14; -0,02], p = 0.0073); ES9: diffg1&g2=-0,08 (95% CI [-0,14; -0,02], p = 0.0068); ES10: diffg1&g2=-0,09 (95% CI [-0,15; -0,02], p = 0.0049); ES7: diffg1&g2=-0,05 (95% CI [-0,10; 0,00], p = 0.0384); ES16: diffg1&g2=-0,06 (95% CI [-0,12; 0,00], p = 0.0403); ES17: diffg1&g2=-0,08 (95% CI [-0,14; -0,02], p = 0.0143); ES18: diffg1&g2=-0,05 (95% CI [-0,10; 0,00], p = 0.0358); | M = 11.26 Mdn=11 SD=2.39 | M = 0.55 Mdn=0.55 SD=0.03 | M = 0.73 Mdn=0.72 SD=0.02 | M = 0.59 Mdn=0.59 SD=0.01 | M = 0.69 Mdn=0.69 SD=0.02 | 0.61 | 0.08 |
BQ1, three groups: x < 6 (n1= 218), 6<=x < 8 (n2=207), x>=8 (n3=248) | ES5: diffg1&g3=-0,01 (95% CI [-0,05; 0,03], p = 0.446), diffg1&g2=-0,05 (95% CI [-0,09; 0,00], p = 0.054), diffg2&g3=0,04 (95% CI [0,00; 0,08], p = 0.102), pg1&g2&g3=0.0449; ES8: diffg1&g3=-0,08 (95% CI [-0,15; -0,02], p = 0.067), diffg1&g2=-0,07 (95% CI [-0,14; 0,00], p = 0.067), diffg2&g3=-0,01 (95% CI [-0,07; 0,06], p = 0.929), pg1&g2&g3=0.0489; ES9: diffg1&g3=-0,09 (95% CI [-0,16; -0,02], p = 0.050), diffg1&g2=-0,08 (95% CI [-0,15; 0,00], p = 0.058), diffg2&g3=-0,01 (95% CI [-0,08; 0,06], p = 0.891), pg1&g2&g3=0.0355; ES11: diffg1&g3=0,08 (95% CI [0,02; 0,14], p = 0.015), diffg1&g2=0,01 (95% CI [-0,05; 0,07], p = 0.858), diffg2&g3=0,07 (95% CI [0,02; 0,13], p = 0.015), pg1&g2&g3=0.0108; | M = 4.85 Mdn=4 SD=1.8 | M = 1.02 Mdn=1.03 SD=0.03 | M = 0.48 Mdn=0.47 SD=0.03 | M = 1.06 Mdn=1.06 SD=0.01 | M = 0.40 Mdn=0.40 SD=0.02 | 0.37 | 0.03 |
BQ2, two groups: x < 2 (n1=219), x>=2 (n2=454) | ES11: diffg1g2=-0,08 (95% CI [-0,13; -0,03], p = 0.0014); ES6: diffg1g2=-0,06 (95% CI [-0,10; -0,02], p = 0.0039); ES3: diffg1g2=0,06 (95% CI [0,01; 0,12], p = 0.0476); ES14: diffg1g2=0,06 (95% CI [-0,01; 0,12], p = 0.04); ES15: diffg1g2=0,08 (95% CI [0,01; 0,14], p = 0.0189); | M = 5.55 Mdn=6 SD=2.91 | M = 0.57 Mdn=0.56 SD=0.04 | M = 0.69 Mdn=0.69 SD=0.02 | M = 0.63 Mdn=0.63 SD=0.01 | M = 0.66 Mdn=0.66 SD=0.02 | 0.67 | -0.01 |
BQ4, two groups: x < 2 (n1=364), x>=2 (n2=309) | ES6: diffg1g2=-0,06 (95% CI [-0,09; -0,02], p = 0.0064); ES11: diffg1g2=-0,06 (95% CI [-0,10; -0,01], p = 0.0165); | M = 3.44 Mdn=3 SD=1.28 | M = 0.67 Mdn=0.67 SD=0.01 | M = 0.59 Mdn=0.60 SD=0.02 | M = 0.68 Mdn=0.67 SD=0 | M = 0.57 Mdn=0.57 SD=0.02 | 0.54 | 0.03 |
BQ5, two groups: x < 7 (n1=274), x>=7 (n2=399) | ES6: diffg1g2=0,06 (95% CI [0,02; 0,10], p = 0.0024); ES9: diffg1g2=-0,08 (95% CI [-0,14; -0,02], p = 0.0043); ES10: diffg1g2=-0,08 (95% CI [-0,15; -0,02], p = 0.0036); ES11: diffg1g2=0,06 (95% CI [0,01; 0,11], p = 0.0168); ES16: diffg1g2=-0,06 (95% CI [-0,12; -0,01], p = 0.0271); ES17: diffg1g2=-0,07 (95% CI [-0,13; -0,01], p = 0.0303); | M = 3.27 Mdn=3 SD=1.65 | M = 0.64 Mdn=0.63 SD=0.02 | M = 0.65 Mdn=0.65 SD=0.03 | M = 0.66 Mdn=0.67 SD=0 | M = 0.60 Mdn=0.60 SD=0.02 | 0.59 | 0,01 |
BQ5, three groups: x < 6 (n1=190), 6<=x < 8 (n2=271), x>=8 (n3=212) | ES9: diffg1&g3=-0,15 (95% CI [-0,22; -0,07], p = 0.0005), diffg1&g2=-0,09 (95% CI [-0,17; -0,02], p = 0.0230), diffg2&g3=-0,05 (95% CI [-0,12; 0,02], p = 0.0965), pg1&g2&g3=0.0007; ES10: diffg1&g3=-0,15 (95% CI [-0,23; -0,07], p = 0.0004), diffg1&g2=-0,09 (95% CI [-0,17; -0,02], p = 0.0112), diffg2&g3=-0,06 (95% CI [-0,13; 0,02], p = 0.1699), pg1&g2&g3=0.0005; ES6: diffg1&g3=0,07 (95% CI [0,03; 0,12], p = 0.016), diffg1&g2=0,02 (95% CI [-0,02; 0,07], p = 0.393), diffg2&g3=0,05 (95% CI [0,01; 0,09], p = 0.023), pg1&g2&g3=0.0093; ES8: diffg1&g3=-0,11 (95% CI [-0,18; -0,04], p = 0.013), diffg1&g2=-0,09 (95% CI [-0,16; -0,02], p = 0.013), diffg2&g3=-0,02 (95% CI [-0,08; 0,05], p = 0.985), pg1&g2&g3=0.0117; ES16: diffg1&g3=-0,09 (95% CI [-0,16; -0,02], p = 0.04), diffg1&g2=-0,08 (95% CI [-0,15; -0,01], p = 0.04), diffg2&g3=-0,01 (95% CI [-0,08; 0,05], p = 0.76), pg1&g2&g3=0.0301; ES17: diffg1&g3=-0,10 (95% CI [-0,18; -0,03], p = 0.04), diffg1&g2=-0,08 (95% CI [-0,15; -0,01], p = 0.06), diffg2&g3=-0,02 (95% CI [-0,09; 0,04], p = 0.53), pg1&g2&g3=0.0329; ES20: diffg1&g3=0,04 (95% CI [-0,02; 0,09], p = 0.022), diffg1&g2=-0,01 (95% CI [-0,07; 0,04], p = 0.928), diffg2&g3=0,05 (95% CI [0,00; 0,10], p = 0.022), pg1&g2&g3=0.0139; | M = 3.63 Mdn=4 SD=1.33 | M = 1.05 Mdn=1.05 SD=0.02 | M = 0.44 Mdn=0.44 SD=0.03 | M = 1.07 Mdn=1.07 SD=0.01 | M = 0.42 Mdn=0.43 SD=0.03 | 0.40 | 0.02 |
BQ6, two groups: x < 7 (n1=318), x>=7 (n2=355) | ES11: diffg1&g2=0,08 (95% CI [0,04; 0,13], p = 0.0006); ES6: diffg1&g2=0,06 (95% CI [0,02; 0,09], p = 0.0056); | M = 5.35 Mdn=5 SD=1.61 | M = 0.62 Mdn=0.63 SD=0.02 | M = 0.63 Mdn=0.63 SD=0.03 | M = 0.65 Mdn=0.65 SD=0 | M = 0.60 Mdn=0.60 SD=0,02 | 0.53 | 0.07 |
BQ6, three groups: x < 6 (n1=240), 6<=x < 8 (n2=229), x>=8 (n3=204) | ES11: diffg1&g3=0,09 (95% CI [0,03; 0,15], p = 0.0077), diffg1&g2=0,03 (95% CI [-0,03; 0,08], p = 0.3516), diffg2&g3=0,06 (95% CI [0,00; 0,12], p = 0.0649), pg1&g2&g3=0.0098; ES6: diffg1&g3=0,07 (95% CI [0,02; 0,12], p = 0.019), diffg1&g2=0,04 (95% CI [0,00; 0,09], p = 0.141), diffg2&g3=0,03 (95% CI [-0,02; 0,07], p = 0.141), pg1&g2&g3=0.0199; | M = 3.89 Mdn=4 SD=1.8 | M = 1.05 Mdn=1.06 SD=0.03 | M = 0.41 Mdn=0.41 SD=0.03 | M = 1.08 Mdn=1.08 SD=0 | M = 0.39 Mdn=0.39 SD=0.03 | 0.36 | 0.03 |
BQ7, two groups: x < 7 (n1=201), x>=7 (n2=472) | ES6: diffg1&g2=0,08 (95% CI [0,04; 0,12], p = 0.0005); ES11: diffg1&g2=0,07 (95% CI [0,02; 0,12], p = 0.0078); ES19: diffg1&g2=0,07 (95% CI [0,02; 0,11], p = 0.0048); | M = 7.26 Mdn=7 SD=1.63 | M = 0.53 Mdn=0.54 SD=0.02 | M = 0.75 Mdn=0.74 SD=0.01 | M = 0.59 Mdn=0.59 SD=0 | M = 0.72 Mdn=0.72 SD=0.01 | 0.70 | 0.02 |
BQ7, three groups: x < 6 (n1=143), 6<=x < 8 (n2=214), x>=8 (n3=316) | ES6: diffg1&g3=0,09 (95% CI [0,04; 0,13], p = 0.0051), diffg1&g2=0,04 (95% CI [-0,01; 0,09], p = 0.1801), diffg2&g3=0,05 (95% CI [0,00; 0,09], p = 0.0619), pg1&g2&g3=0.0042; ES11: diffg1&g3=0,10 (95% CI [0,04; 0,16], p = 0.0086), diffg1&g2=0,03 (95% CI [-0,04; 0,09], p = 0.4526), diffg2&g3=0,07 (95% CI [0,02; 0,12], p = 0.0186), pg1&g2&g3=0.0035; | M = 1.31 Mdn=1 SD=0.61 | M = 1.05 Mdn=1.06 SD=0.02 | M = 0.45 Mdn=0.45 SD=0.02 | M = 1.07 Mdn=1.07 SD=0.01 | M = 0.47 Mdn=0.48 SD=0.02 | 0.47 | 0.00 |
BQ8, two groups: x < 2 (n1=123), x>=2 (n2=550) | ES4: diffg1&g2=-0,11 (95% CI [-0,17; -0,05], p = 0.0001); ES12: diffg1&g2=-0,13 (95% CI [-0,20; -0,06], p = 0.0002); ES14: diffg1&g2=-0,20 (95% CI [-0,27; -0,13], p = 0.0000); ES15: diffg1&g2=-0,20 (95% CI [-0,28; -0,12], p = 0.0000); ES3: diffg1&g2=-0,10 (95% CI [-0,16; -0,03], p = 0.0031); ES10: diffg1&g2=-0,12 (95% CI [-0,20; -0,04], p = 0.0050); ES11: diffg1&g2=-0,08 (95% CI [-0,14; -0,02], p = 0.0058); ES8: diffg1&g2=-0,09 (95% CI [-0,16; -0,02], p = 0.0225); ES9: diffg1&g2=-0,10 (95% CI [-0,18; -0,03], p = 0.0142); ES13: diffg1&g2=-0,07 (95% CI [-0,14; -0,01], p = 0.0223); ES16: diffg1&g2=-0,10 (95% CI [-0,17; -0,02], p = 0.0159); ES17: diffg1&g2=-0,09 (95% CI [-0,17; -0,02], p = 0.0319); ES18: diffg1&g2=-0,08 (95% CI [-0,14; -0,01], p = 0.0242); | M = 6.14 Mdn=6 SD=1.69 | M = 0.42 Mdn=0.42 SD=0.02 | M = 0.83 Mdn=0.83 SD=0.01 | M = 0.48 Mdn=0.48 SD=0.01 | M = 0.79 Mdn=0.78 SD=0.01 | 0.82 | -0.03 |
BQ9, two groups: x < 51 (n1=333), x>=51 (n2=340) | ES1: diffg1&g2=0,09 (95% CI [0,05; 0,12], p = 0.0000); ES2: diffg1&g2=0,10 (95% CI [0,06; 0,13], p = 0.0000); ES3: diffg1&g2=0,13 (95% CI [0,08; 0,18], p = 0.0000); ES4: diffg1&g2=0,10 (95% CI [0,05; 0,15], p = 0.0006); ES5: diffg1&g2=0,08 (95% CI [0,04; 0,11], p = 0.0000); ES14: diffg1&g2=0,14 (95% CI [0,08; 0,19], p = 0.0001); ES15: diffg1&g2=0,14 (95% CI [0,08; 0,20], p = 0.0001); ES7: diffg1&g2=0,06 (95% CI [0,01; 0,10], p = 0.0133); ES8: diffg1&g2=0,09 (95% CI [0,04; 0,15], p = 0.0466); ES11: diffg1&g2=-0,05 (95% CI [-0,09; 0,00], p = 0.0485); ES13: diffg1&g2=0,05 (95% CI [0,01; 0,10], p = 0.0297); ES19: diffg1&g2=0,04 (95% CI [0,00; 0,08], p = 0.0193); | M = 5.79 Mdn=6 SD=1.44 | M = 0.58 Mdn=0.58 SD=0.02 | M = 0.69 Mdn=0.69 SD=0.02 | M = 0.61 Mdn=0.61 SD=0.01 | M = 0.68 Mdn=0.68 SD=0.02 | 0.51 | 0.17 |
BQ9, three groups: x < 40 (n1=225), 40<=x < 60 (n2=231), x>=60 (n3=217) | ES1: diffg1&g3=0,10 (95% CI [0,06; 0,14], p = 0.0000), diffg1&g2=0,07 (95% CI [0,03; 0,11], p = 0.0002), diffg2&g3=0,03 (95% CI [-0,01; 0,06], p = 0.0716), pg1&g2&g3=0.0000; ES2: diffg1&g3=0,12 (95% CI [0,08; 0,16], p = 0.0000), diffg1&g2=0,07 (95% CI [0,03; 0,11], p = 0.0007), diffg2&g3=0,05 (95% CI [0,01; 0,09], p = 0.0162), pg1&g2&g3=0.0000; ES3: diffg1&g3=0,17 (95% CI [0,11; 0,23], p = 0.0000), diffg1&g2=0,06 (95% CI [0,01; 0,12], p = 0.110), diffg2&g3=0,10 (95% CI [0,04; 0,17], p = 0.003), pg1&g2&g3=0.0000; ES4: diffg1&g3=0,13 (95% CI [0,07; 0,19], p = 0.0011), diffg1&g2=0,02 (95% CI [-0,04; 0,07], p = 0.5450), diffg2&g3=0,11 (95% CI [0,05; 0,18], p = 0.0011), pg1&g2&g3=0.0004; ES5: diffg1&g3=0,09 (95% CI [0,04; 0,13], p = 0.0000), diffg1&g2=0,03 (95% CI [-0,01; 0,08], p = 0.042), diffg2&g3=0,05 (95% CI [0,01; 0,10], p = 0.012), pg1&g2&g3=0.0000; ES14: diffg1&g3=0,17 (95% CI [0,11; 0,24], p = 0.0002), diffg1&g2=0,05 (95% CI [-0,01; 0,12], p = 0.6097), diffg2&g3=0,12 (95% CI [0,05; 0,19], p = 0.0031), pg1&g2&g3=0.0002; ES15: diffg1&g3=0,18 (95% CI [0,11; 0,25], p = 0.0004), diffg1&g2=0,06 (95% CI [0,00; 0,13], p = 0.5430), diffg2&g3=0,12 (95% CI [0,04; 0,19], p = 0.0082), pg1&g2&g3=0.0006; ES7: diffg1&g3=0,09 (95% CI [0,03; 0,15], p = 0.0064), diffg1&g2=0,01 (95% CI [-0,04; 0,06], p = 0.7293), diffg2&g3=0,08 (95% CI [0,02; 0,14], p = 0.0120), pg1&g2&g3=0.0043; ES11: diffg1&g3=-0,08 (95% CI [-0,14; -0,03], p = 0.0069), diffg1&g2=-0,10 (95% CI [-0,15; -0,04], p = 0.0033), diffg2&g3=0,02 (95% CI [-0,04; 0,07], p = 0.5752), pg1&g2&g3=0.0017; ES8: diffg1&g3=0,13 (95% CI [0,06; 0,20], p = 0.016), diffg1&g2=0,03 (95% CI [-0,04; 0,09], p = 0.956), diffg2&g3=0,11 (95% CI [0,04; 0,18], p = 0.016), pg1&g2&g3=0.0116; ES10: diffg1&g3=0,14 (95% CI [0,07; 0,22], p = 0.034), diffg1&g2=0,02 (95% CI [-0,05; 0,09], p = 0.995), diffg2&g3=0,12 (95% CI [0,04; 0,20], p = 0.034), pg1&g2&g3=0.0245; ES19: diffg1&g3=0,06 (95% CI [0,01; 0,11], p = 0.034), diffg1&g2=0,03 (95% CI [-0,01; 0,08], p = 0.198), diffg2&g3=0,03 (95% CI [-0,03; 0,08], p = 0.198), pg1&g2&g3=0.0351; ES20: diffg1&g3=-0,07 (95% CI [-0,12; -0,02], p = 0.127), diffg1&g2=0,01 (95% CI [-0,04; 0,06], p = 0.268), diffg2&g3=-0,08 (95% CI [-0,13; -0,02], p = 0.026), pg1&g2&g3=0.0253; | M = 7.13 Mdn=7 SD=1.45 | M = 0.93 Mdn=0.93 SD=0.03 | M = 0.54 Mdn=0.54 SD=0.03 | M = 0.98 Mdn=0.98 SD=0.01 | M = 0.50 Mdn=0.50 SD=0.03 | 0.34 | 0.16 |
2. Guidance about giving the “need for help” ratings for expression statements
3. Formulation of the questionnaire items
4. Formulation of machine learning experiments
Model: “sequential” Parameters: total 73,112; trainable: 73,112; non-trainable: 0 | ||
---|---|---|
Layer (type)
|
Output shape
|
Number of parameters
|
rescaling_1 (Rescaling) | (None, 20, 25, 3) | 0 |
conv2d (Conv2D) | (None, 20, 25, 16) | 448 |
max_pooling2d (MaxPooling2D) | (None, 10, 12, 16) | 0 |
conv2d_1 (Conv2D) | (None, 10, 12, 32) | 4640 |
max_pooling2d_1 (MaxPooling2D) | (None, 5, 6, 32) | 0 |
conv2d_2 (Conv2D) | (None, 5, 6, 64) | 18,496 |
max_pooling2d_2 (MaxPooling2D) | (None, 2, 3, 64) | 0 |
flatten (Flatten) | (None, 384) | 0 |
dense (Dense) | (None, 128) | 49,280 |
dense_1 (Dense) | (None, 2) | 258 |
Results
1. Addressing the main research question RQ1
1.1 Identifying statistically significant rating differences for expression statements in respect to background questions
1.1.1 Participants and stages
1.1.2 Descriptive data
Background question (BQ)
|
Answer value
| |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
1
|
2
|
3
|
4
|
5
|
6
|
7
|
8
|
9
|
M
|
Mdn
|
SD
| |
BQ1: an estimated health condition | 11 (2%) | 3 (~0%) | 40 (6%) | 66 (10%) | 98 (15%) | 45 (7%) | 162 (24%) | 129 (19%) | 119 (18%) | 6.53 | 7 | 1.97 |
BQ5: the quality of life | 7 (1%) | 6 (1%) | 29 (4%) | 47 (7%) | 101 (15%) | 84 (12%) | 187 (28%) | 123 (18%) | 89 (13%) | 6.53 | 7 | 1.77 |
BQ6: the satisfaction about health | 17 (3%) | 10 (1%) | 68 (10%) | 61 (9%) | 84 (12%) | 78 (12%) | 151 (22%) | 138 (21%) | 66 (10%) | 6.13 | 7 | 2.04 |
BQ7: the satisfaction about ability | 8 (1%) | 9 (1%) | 44 (7%) | 28 (4%) | 54 (8%) | 58 (9%) | 156 (23%) | 128 (19%) | 188 (28%) | 6.98 | 7 | 1.98 |
Background question (BQ)
|
Answer value
|
---|---|
BQ2: a health problem reduces ability | No (coded as 1): 219 (33%); Yes (coded as 2): 454 (67%) (M = 1.67; Mdn=2; SD=0.47) |
BQ3: one or more diseases identified by a doctor | Disease category (the number of unique persons who selected the category): Lung diseases: 126; Heart and circulatory diseases: 177; Joint and back diseases: 301; Injuries:103; Mental health problems: 188; Vision and hearing deficits: 191; Other diseases: 345 |
BQ4: a continuous or repeated need for a doctor’s care | No (coded as 1): 364 (54%); Yes (coded as 2): 309 (46%) (M = 1.46; Mdn=1; SD=0.50) |
BQ8: the sex | Man (coded as 1): 123 (18%); Woman (coded as 2): 550 (82%) (M = 1.82; Mdn=2; SD=0.39) |
BQ9: the age | Belonging to an age range category (the lower bound is included in the range but not the upper bound): 16-20 years: 143 (21%); 20-30 years: 21 (3%); 30-40 years: 61 (9%); 40-50 years: 96 (14%); 50-60 years: 135 (20%); 60-70 years: 141 (21%); 70-80 years: 64 (10%); 80-90 years: 12 (2%); 90 years or more: 0 (0%) (M = 46.93; Mdn=51; SD=19.57) |
1.1.3 Outcome data, main results and other analyses
A pair of expression statements (ES) and background questions (BQ)
|
Kendall rank-correlation measure
|
Cosine similarity measure
|
---|---|---|
ES16&ES17 | 0.91 | 0.97 |
ES14&ES15 | 0.86 | 0.95 |
ES9&ES10 | 0.79 | 0.92 |
ES16&ES18 | 0.78 | 0.90 |
ES17&ES18 | 0.77 | 0.89 |
ES7&ES8 | 0.75 | 0.87 |
ES1&ES2 | 0.73 | 0.80 |
BQ1&BQ6 | 0.71 | 0.82 |