Dr. Benhima has a PhD in education. He is specialized in data analysis and research design. He currently works as an assistant professor at Mohammed V University in Rabat, Morocco.
Thanks for such insightful automatic lists of tables and lists of figures, in my Master's thesis I did them manually, and now I know how to do them automatically, thank you!!!
Hi sir. Thank you very much for the great explanation. I have one question. Is it possible to use three scale likert (namely, True, False, I dont know) in SmartPls? Thank you very much in advance.
Hello sir, I have a doubt. What is the best way to test normality for likert scale data shapiro wilk or kurtosis & skewness . What test we should use. Also, if your data is normal according to kurtosis & skewness test but indicates non- normality according to shapiro wilk. What should we report and which one is more reliable ? Sir please do reply
Hello Sir a warm Greetings from my side. Firstly I wanna say that amazin video with great exlaination. It helped me alot as im doing my final year MBA project report. May I have your insta handle, because I need to get clearity on my analysis part in project report.
The data was checked its normality by using SPSS. Then, the results is not normal, so the data was transformed and became normal. To run dependent T-test, should i use the actual data or transformed data?
thank you very much. would clarify the first steps in SPSS you mentioned in the video before conducting regression and correlation analysis. e.g. how do you calculate total scores of each variable that have several dimensions with several items for each dimension using SPSS? thanks. For instance, suppose we have 3 main variables (independent variable), (dependent Variable) and (Mediating variable). each variable is measured by some already established and validated scale. and each scale has about 4 Dimensions with about 3 statements for each dimension. and all items were scored on a 7-point Likert scale 1=Strongly Disagree to 7=Strongly Agree. How would be the correct analysis process for each dimension and each scale? and how would be the correlation analysis between the 3 variables? and for the total scores of dimensions/scales, should we use averages (means) or sums and why?. thank you very much in advanced for your kind clarification.
Thank you Prof. suppose we have 3 main variables (independent variable), (dependent Variable) and (Mediating variable). each variable is measured by some already established and validated scale. and each scale has about 4 Dimensions with about 3 statements for each dimension. and all items were scored on a 7-point Likert scale 1=Strongly Disagree to 7=Strongly Agree. How would be the correct analysis process for each dimension and each scale? and how would be the correlation analysis between the 3 variables? and for the total scores of dimensions/scales, should we use averages (means) or sums and why?. thank you very much in advanced for your kind clarification.
Thank you very much. suppose we have 3 main variables (independent variable), (dependent Variable) and (Mediating variable). each variable is measured by some already established and validated scale. and each scale has about 4 Dimensions with about 3 statements for each dimension. and all items were scored on a 7-point Likert scale 1=Strongly Disagree to 7=Strongly Agree. How would be the correct analysis process for each dimension and each scale? and how would be the correlation analysis between the 3 variables? and for the total scores of dimensions/scales, should we use averages (means) or sums and why?. thank you very much in advanced for your kind clarification.
Good day! Likert scale itself is inherently not normal (Aboy, 2023). The appropriate measure of average and dispersion is Median and IQR. Median is less sensitive to outliers and skewed distributions compared to mean. IQR is less influenced by extreme values compared to SD.