so on SPSS I did the kolmogorov smirnov test to see if my data has normal distribution but the thing is I didn't specify what my missing data was as in I just opened the file on SPSS by selecting the excel file. Is that okay?
If your data is normally distributed without removing outliers, then it should be reported that way. Missing data needs to be explained. Is this a single point analysis (cross-sectional)? If so, missing data needs to be examined as to the percent per variable and also if there is a trend. For example, reporting income may be a sensitive issue and participants may omit answering the question. There could be a bias for those who have a low income being more likely to omit responding compared to mid-income or high-income. Data set can be imported from Excel to SPSS but they need to be cleaned. This means independent researchers choose to compare 10% of the questionnaire or laboratory report to the transcribed data to insure accuracy. Another form of missing data is the response ' not sure' or 'I don't know' If the percent of answers 'do not know' is 2% or less, you could be justified in stating "We assigned 'I do not know" responses as missing since there were
SPSS can tell that your data is missing if you have blanks. But also, if you have a large sample size, the kolmogorov smirnov test will always be significant. I would use another method to check your distributional assumptions...
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS