Sir, one more question please.
If I want to know significant difference between my three groups on individual 6 items( which are measured on Likert 5 point scale), which form a construct, can I use One way ANOVA by doing it individually on all 6 items of the construct, or shall I combine all 6 to a means score and then analyse the difference between the groups on the construct using one way ANOVA. If I follow this second route, then can I use descriptive mean score of the individual items to say which individual items is at a higher level & which is a lower level between the three groups ?
Or is there any other more appropriate way?
I know Likert items is ordinal, but I have read that if it has atleast 5 points of measurement, then it may be treated as interval data & parametric tests can be conducted on it. I don't want to use non parametric Kruskal wallis test as it is weaker.
To analyze Likert items, typically used in surveys to measure attitudes or opinions, start by summarizing the data using descriptive statistics such as means, medians, and mode to understand central tendencies. Frequencies and percentages can also be useful to show the distribution of responses across categories. For deeper insights, inferential statistics like chi-square tests, t-tests, or ANOVA can be employed to compare groups or check for significant differences if the data meets necessary assumptions. Additionally, factor analysis may be used to explore underlying dimensions in the data, especially with multiple Likert items meant to measure similar constructs. Visualization tools like bar charts or histograms are effective in presenting the distribution of responses, helping to illustrate patterns or trends in the data.
Karan Berry it is important to draw a distinction between the item ratings and the overall scale. The 5-point item rating structure produce ordinal level scores. However, the summation of the item rating scores is what produces the scale developed by Professor Likert. While it is tempting to focus on the item results, the power of the technique lies in the accumulation of the total item scores. This innovation allows for the accumulation of information at the item level to produce a more robust understanding of the underlying construct of interest.
The total score (the mean is just an arithmetic transformation of the total score) is what you would use to understand the differences amongst the groups based on their scores on the instrument.
- Badges
- Science topic