Thank you for this great video. I have a couple of questions. 1) What if we have more than 20% of the data for one variable missing? 2) What if all missing data are from one case. Drop it from analysis? And 3) What if the results of t-tests come back statistically significant? Thanks!
@Michelle Hollenbaugh Hi Michelle! Nice video🙂 I wish to know how to calculate the reliability of a questionnaire in spss if we have missing data in some questions? Is it using the usual cronbach alpha?? And what should be putten in the cells of the missing data in spss?
I noticed that when you recoded rincom91 into a different variable rincom91_md and changed old values of "system or user-missing" to 0 and all other values to 1 that the values of 0 in rincom91 were recoded as 0. But should they not be a 1? It is not a missing value in the same way an empty cell is. What if 0 IS an answer of that participant's rincom91? Would that not affect the things done later?
Hey! This was SO helpfull, thank you!! I have a few questions. I am studying the effect of SES, IQ, school achievement, and teacher expectations on academic self concept. I have missing data in all my independent variables. So do i have to make a 'missing data' variable for all of them? And than use these in my mediation analysis? Thanks!
Hi, when you check that the difference is significant you were actually checking the P value for the Levene's test for equality of variance. Shouldn't you be looking at the p value for mean differences?
OK, that was fantastic!! I couldn't even expect it to be so complicated when my supervisor asked me to use the sample mean for the missing answers! Just one question if I may, I have a few respondents who haven't answered 1 question out of the 3 or for that there are in the questionnaire. Is it advisable to follow the procedure you described or shall I remove them completely?
hi! i love this tutorial and it really helped me out so thank you! but question, if for example, my mean is 2.93, do i round up so it'll be 3 or do i just take the 2 to use as my missing variable? thanks!
Hi Katrina! Thanks! This would be based on your judgement- I personally would probably round up to 3, but I think you also need to consider your variable, and what the difference between a '2' and a '3' is on your measure; and how that relates to your sample. I hope that helps!
@Michelle I got an 'insufficient memory to complete the operation in SPSS' after i used the multiple imputation technique. wht can i do? any suggestion to improve memory. thanks
Thank you so much ma'am, you saved me as I didn't know what to do about the missing data nor understand where they came from.
Currently working on my thesis and finding this incredibly helpful. Have a NICE day!
This is great!! Thank you! Please continue uploading these videos :) You are a great stats instructor!
okie dokie artichokieeee..
THANK YOU, THANK YOU, you have blessed me, and saved me for this class
You are brilliant and just made my day! So easy to follow and incredibly helpful!!! Thank you!
Thank you for this great video. I have a couple of questions. 1) What if we have more than 20% of the data for one variable missing? 2) What if all missing data are from one case. Drop it from analysis? And 3) What if the results of t-tests come back statistically significant? Thanks!
hi, did you get the answer for 2)?
This was so helpful and so easy to follow.
Michelle, this was SUPER helpful! Thank you!
Extremely helpful! Thank you!!
@Michelle Hollenbaugh Hi Michelle! Nice video🙂 I wish to know how to calculate the reliability of a questionnaire in spss if we have missing data in some questions? Is it using the usual cronbach alpha?? And what should be putten in the cells of the missing data in spss?
I noticed that when you recoded rincom91 into a different variable rincom91_md and changed old values of "system or user-missing" to 0 and all other values to 1 that the values of 0 in rincom91 were recoded as 0. But should they not be a 1? It is not a missing value in the same way an empty cell is. What if 0 IS an answer of that participant's rincom91? Would that not affect the things done later?
Hey! This was SO helpfull, thank you!! I have a few questions. I am studying the effect of SES, IQ, school achievement, and teacher expectations on academic self concept. I have missing data in all my independent variables. So do i have to make a 'missing data' variable for all of them? And than use these in my mediation analysis? Thanks!
Nice video ❤😂🎉😢
Love the personality, very informational.
Hi, when you check that the difference is significant you were actually checking the P value for the Levene's test for equality of variance. Shouldn't you be looking at the p value for mean differences?
My thoughts exactly
Thank you for this very helpful video! Do you have any literature that could be used as reference for this procedure? 😇 Thanks in advance!
Excuse me,
Can anyone tell me whether the skipped questions are considered to be the missing data or not?
OK, that was fantastic!! I couldn't even expect it to be so complicated when my supervisor asked me to use the sample mean for the missing answers! Just one question if I may, I have a few respondents who haven't answered 1 question out of the 3 or for that there are in the questionnaire. Is it advisable to follow the procedure you described or shall I remove them completely?
How can we perform predictive mean matching in SPSS and extract pooled imputed data set for machine learning in other tools such as python
You just saved my life.
hi! i love this tutorial and it really helped me out so thank you! but question, if for example, my mean is 2.93, do i round up so it'll be 3 or do i just take the 2 to use as my missing variable? thanks!
Hi Katrina! Thanks! This would be based on your judgement- I personally would probably round up to 3, but I think you also need to consider your variable, and what the difference between a '2' and a '3' is on your measure; and how that relates to your sample. I hope that helps!
yep! i get it now! thank you so much for replying! have good day ^^
Thank u so much for this!
Hi... Can we do the same if data is not distributed normally ???
@Michelle I got an 'insufficient memory to complete the operation in SPSS' after i used the multiple imputation technique. wht can i do? any suggestion to improve memory. thanks
Hi! You know, I'm not sure, that sounds like a problem with the software, you may want to contact IBM. Sorry I'm not more of a help!
booming 🎉
Thank you!
*okie dokie artichokie*
Thank you👍
Thanks michelle
Not working for my string data
what about categorical data?
nice one
How did you do it?? You should have made this video from the start. How did you even find the table of missing value?? This is so confusing.
Does not tell how to find that missing data my professor wants the missing data
Hoe doe je in een keer alle variabelen kopiëren broer???
control A maat
i found this interesting