0

There was a study done on the doctors whose knowledge is tested on the laws towards the patients. Data are collected using a questionnaire before and after giving intervention. Each question have options as

  • 1 = `Strongly disagree'
  • 2 = 'Disagree'
  • 3 = 'Neutral'
  • 4 = 'Agree'
  • 5 ='Strongly agree

How to compare the data and answer the following question:

Is there a significant difference in knowledge before and after intervention?

Paired Sample T-test can be applied?? (I strongly doubt because this data does not follow normal)

What about Signed Rank test??

A model data can be created as follows:

n=50

df = data.frame(id =c(seq(1,n),seq(1,n)), pre_post = c(rep(0,n),rep(1,n)), q1 = sample(1:5,2*n, replace = TRUE), q2 = sample(1:5,2*n, replace = TRUE),q3 = sample(1:5,2*n, replace = TRUE), q4 = sample(1:5,2*n, replace = TRUE))

df$pre_post = factor(df$pre_post, levels = c(0,1), labels = c('Pre', 'Post'))



df$pre_post = as.factor(df$pre_post)
df$q1 = ordered(df$q1, levels =c(1,2,3,4,5), labels = c('Strongly disagree', 'Disagree', 'Neutral', 'Agree', 'Strongly agree'))
df$q2 = ordered(df$q2,levels =c(1,2,3,4,5), labels = c('Strongly disagree', 'Disagree', 'Neutral', 'Agree', 'Strongly agree'))
df$q3 = ordered(df$q3,levels =c(1,2,3,4,5), labels = c('Strongly disagree', 'Disagree', 'Neutral', 'Agree', 'Strongly agree'))
df$q4 = ordered(df$q4,levels =c(1,2,3,4,5), labels = c('Strongly disagree', 'Disagree', 'Neutral', 'Agree', 'Strongly agree'))

enter image description here

gloom
  • 101

1 Answers1

0

Wilcoxon signed-rank W test takes data, whether it is non-normal real or Likert scale, and converts it into approximately normal ranked data. The W-statistic is then calculated and from that a probability of data identical location (location is more general than mean or median) is calculated.

An example of R-language application is given here and here, and a video is here.

Carl
  • 13,084