Skip to main content

Table 1. Demographic information of participants.

From: For Better or Worse: The Impact of Counterfactual Explanations’ Directionality on User Behavior in xAI

 

Before quality assurance measures (N = 161)

 

control

upward

downward

mixed

H value\(^a\)

p value

N

40

40

40

41

  

Gender\(^{b,c}\)

20f/20m

20f/20m

20f/20m

20f/19m/1nb

0.018

.999

Age (Mdn)\(^{b,d}\)

35–44y

35–44y

25–34y

25–34y

5.091

.165

 

After quality assurance measures (N = 158)

 

control

upward

downward

mixed

H value\(^a\)

p value

N

40

39

38

41

  

Gender\(^{b,c}\)

20f/20m

20f/19m

19f/19m

20f/19m/1nb

0.214

.975

Age (Mdn)\(^{b,d}\)

35–44y

35–44y

25–34y

25–34y

7.388

.061

  1. \(^a\) non-parametric Kruskal-Wallis H test
  2. \(^b\) note that one participant did not disclose age or gender information
  3. \(^c\) f = female, m = male, nb = non-binary/gender non-conforming
  4. \(^d\) Mdn = median age (options: 18–24y, 25–34y, 35–44y, 45–54y, 55–64y, 65y and over).