Replacing Missing Values in SPSS with the Series Mean

[@GirlAlmighty93x]

Thank you Dr. Todd Grande, you have taught me more in 12 1/2 minutes than my tutor in one semester! With your support, I was able to fix up my data sheet and complete my analysis. Keep up the good work!

[@cheali5857]

very well explained, step by step with nothing left out for students to understand and be able to redo the procedures themselves.

[@saifaldeen1716]

Thanks Doctor, Guys just move on to min 10:43 and will find your answer

[@elekfazekas8385]

This is really useful and the pace of explonation is perfect. Thank you so much, you saved me a lot of time! However I'm curious about what to do when the missing values are above or equal to 5%?

[@yardenlevy3182]

Yeah that would be super helpful

[@chuaterence7753]

Dear Dr. Todd, thank you for tutorial. It is very helpful. May I ask what would be an appropriate statistical method to replace missing values if the percentage of missing values is more than 5%?

[@yardenlevy3182]

Yeah that would be super helpful

[@shahinghaziani8504]

That's also my question. Thank you

[@arefameeno]

as i knew, if missing values is more than 5%, u cannt replace with anything as it will effect the result. You just leave them as missing values

[@katy8791]

@@arefameeno Hello! Any idea about 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?

[@rebccajg]

Thanks, Todd. You're such a homie.

[@paulinbill1568]

wow ...wow...wow.....amazing ! to make it short I just remember Transform....replace missing value......select series mean !

[@shahinghaziani8504]

Dear Dr. Grande, Thank you very much for your useful tutorials. I've been learning a lot from you. I just don't understand why is it necessary that we replace the missing cells first with -1? Why not just replacing the missings with series mean right at the beginning?

[@johnburke2047]

Hi, thanks for the helpful explanation. I want to calculate a total score of a questionnaire with a couple of missing items, but want the missing data points to be included as the mean of the specific case/person's scores on that questionnaire only, rather than the dataset mean for that questionnaire. Is there a way to code this... for example SUM.3(Item_1, Item_2, Item_3, Item_4, Item_5) IF SYSMIS, use MEAN) Thanks for your help!

[@piertje5927]

Hi, I know i'm super late but I have the exact same question as you! I'm really curieus if you found out the answer to your question yourself? And what is the answer? Because I don't know how to code this in SPSS as well. Fingers crossed you will answer!

[@curiousstudent1161]

I've replaced the missing values with Series mean(all values expected to be from 1 to 5 on a Likert scale) and instead of whole numbers, I've got 3.77. How do I replace the missing values in a way that I get only 1, 2, 3 4 or 5? Thank you

[@kijafabruce5768]

Go to: variable view > decimal column > change the "2" to "0" (Better late than never).

[@taragibson3655]

@@kijafabruce5768 yes better late than never, you have helped me 2 years later! thank you

[@atakanturkoglu]

Thank you so much for the video Dr. Grande. I was wondering how you were able to default your tables (output) like this. Would you happen to have a video on that?

[@sophievbdk3681]

Does the replacement of missing values with mean or median depend on distribution of data and normality plot?

[@ooaluko]

Thanks for the tutorial. Can I use the resulting data se for multiple regression analysis

[@dr.debpriyokumardey6318]

Thank u Dr. Todd for the elegant explanation. My question is that if I replace the missing values with series mean, and then would want to run some statistic (say univariate ANOVA), then would these series mean values in the different groups not mislead the analysis results? Please respond.

[@linduchyable]

i don't want to replace the missing values i want my negative values as -127.79 to counted in SPSS. my negative values disappeared from the system and the cell still blank. how can i solve this problem??? please help!!

[@ekeneonyenka8236]

hello Dr. Todd Grande, how do I contact you?, I am trying to get the reliability score for a set of data on spss and i got the warning message N=0, i was made to understand the it was because of the missing data set so i tried following your steps to replace the missing data set so i could run the analysis, howevever i keep getting the prompt that the list can only carry a maximum of 100 variables at the stage of replacing the missing values, and i have about 146 variables. kindly assist.

[@heybyrd]

Thanks a million Prof.

[@HannahMarshall-c4m]

Hey there, after you replace missing values, and you get a new variable (say Motivation_1), do you delete the old/original variable Motivation, or just leave it? I can imagine you'll have a s lot of duplicate variables which could get confusing?

[@ShirlyKhoh]

I have the same question too, when running the analysis I assume we will be using the new variables with missing values replaced with the series mean. Since we do not want our data cases to have missing values.

[@boissons999]

Is the series mean calculated excluding missing values lines?

[@dshytikov6311]

Hello! I have the following question: I have a survival study on animals which were fed with special compound in order to increase their lifespan and healthspan. During the experiment we measured a number of parameters (body weight, routine blood tests, physiological performance in some excercises etc). As you may understand, during the experiment mice were dying with more advanced age and with time the amount of missing values increases significantly. How to analyze this kind of data? Simple repeated measure ANOVA uses casewise deletion, but then I can delete all of my data. Mean or LOCF substitutions may affect the data significantly... WHat to do then? Do you guys have any suggestion?

[@katy8791]

Hello Dr. Grande! Nice video! Any idea about 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?

[@ammarbaig2249]

all the missing values which you replaced are same is it possible to replace values by taking mean of before values

[@alexanderfridayeagle9146]

So the mean is 2.51 what do I put in the missing values cage, just round it to 3? I am missing only one item from one questionnaire from one participant.

[@arefameeno]

that is exactky my question. as I use one decimal number while the mean needs at least 2 or more decimal.

[@okikiadeduyite9205]

After the Missing Value has been replaced, If I attempt to get the frequency table, will it still appear?

[@heavenlyblessed2044]

Thanks for the video Dr. Grande. I would like to request a video on missing data using LISTWISE.

[@mienkeknipe9137]

How do I replace missing values not with the Functioning series mean (50) but rather replace missing values for Females with the mean Functioning value of Female & the same applies to Males?

[@iyshak5373]

if I want to conduct a chi square test and there are missing values. how would I report the missing values. will 'N' be a new number..

[@tarahalpern2813]

Thank you for this tutorial, can you replace missing values for binomial/nominal data in the same way as scale values with the series mean?

[@imranaltintop2830]

dear sir, I am writing from Turkey. I have a data completed from 321 samples. ıt consists of STAI, Ways of coping questionnaire and CD-RISC. ı watched your video and others but my mind is confused. everyone says different and complicated things. when I run missing value analysis. I recognize that my item's missing value for all these scales is ranged from 5-20 missing values. then I want to learn that can I use the mean for my missing data? other ways seem to me more complicated.

[@kathystansbury771]

Thank you - why are the Functioning_1 values that replace the missing values in Functioning not the same, if there are replaced with the series mean? It is one mean and the replaced scores should all be the same. This was the case for the motivation variable, but not the functioning variable. I might have missed something....

[@sunnord7634]

Hi! I have coded the missing values to -99, but still cannot make a line graph using repeated measures....with the 6 first measures where all have values it is no problem,, but not with all 11

[@datascientist2958]

How can we implement Predictive mean matching for missing values?

[@johnchoi8643]

Thank you for the video!

[@Itsryanpilot]

Thank you very much for this video. i currently have missing values in my pilot study and i am trying to replace them with the SEM so i can hopefully get a high enough Cr alpha in order for me to do my real study for my thesis.. just to confirm, this way you showed can be used for all cases, am i right?

[@mansoorahmed4150]

I got this error, Can you explain this error related to coding missing data. Warnings!! There are too many cases to fit into memory at one time. If multiple result variables are being generated, you can reduce the memory requirements by breaking up the problem into multiple procedure invocations. Otherwise, use SET WORKSPACE to increase the memory available. Execution of this command stops.

[@BestAmazonReviews723]

can get link to download all SPSS videos DR

[@aqmalliyasaufi3258]

thank you for sharing. it very helpful :)

[@DrGrande]

You're welcome!

[@zoumitechristthierrystephe736]

thank you for video !!!

[@DrGrande]

You're welcome. Thanks for watching.

[@lucialuu2014]

Thank you very much

[@muhammadshahrukh4896]

thankyou

[@Abdul20024]

thank you

[@DrGrande]

You're welcome, thanks for watching -

[@anggeranugerah]

its helpfull

[@DrGrande]

Thank you -

[@linduchyable]

to be counted*

[@sarahlillemo4833]

Whyyyyy no syntax???? Showing a quick paste would make your videos so much more valuable for people trying to get the hang of syntax. Thanks for all the work you put into these videos though

[@lamraouiramzi6913]

thank you