SPSSisFun: Dealing with missing data (Listwise vs Pairwise)

In this video I explain the difference between "excluding cases listwise" and "excluding cases pairwise" when dealing with missing data.
note: excluding cases "analysis by analysis" is the same as excluding cases "pairwise"
If you have any questions please feel free to post them in the comments section below and I will get back to you as soon as I can.
You can also message me on linkedin: / szymanskijason

[@theempress5058]

excellent work ,thanks

[@piusameh102]

Thanks for this!

[@nadak5128]

Thank you

[@jasoncastledine1012]

what if you have multiple independent and dependent variables

[@jasoncastledine1012]

cant fit more than one in the grouping variables ...i have age, gender, school as IVS to 3 different tests on the DV

[@user-rq7so2is6n]

May I ask a further question? In linear regression, if we use listwise deletion, would the models by stepwise, forward selection or backward selection be different?

[@kingsanalytics2193]

In listwise, you should not worry about analysis/outcomes, whether linear regression or hierarchical regression or logistic regression. The basic idea is to remove the entire row/respondent data that is affected by one or more missingness.
Just an addition - you should be fine if the data is missing at random (MAR), ex, due to mistake omission by respondents.
BUT
You may face a problem of low statistical power which leads to invalid conclusion , if the respondent(s) made the omission intentionally (maybe due to the fact that you asked sensitive or ambigious question).

[@sagwatie]

does this mean that for a comparison study for example, observed vs estimated, listwise is the best?

[@kingsanalytics2193]

In listwise, you should not worry about analysis/outcomes, whether linear regression or hierarchical regression or logistic regression. The basic idea is to remove the entire row/respondent data that is affected by one or more missingness.
Just an addition - you should be fine if the data is missing at random (MAR), ex, due to mistake omission by respondents.
BUT
You may face a problem of low statistical power which leads to invalid conclusion , if the respondent(s) made the omission intentionally (maybe due to the fact that you asked sensitive or ambigious question).

[@bassamalsheakhly1889]

Thanks, may i have your e mail? I need help with my data