hi thank you for the video explanation. i want to ask, what if we want to compare a nominal data sample characteristics between 2 groups (nominal vs nominal) ? because in this video only show numerical data (income) vs nominal data (gender). Thank you.
Hello sir, Thank you so much for this video. But I am still confused I want to compare a lot of questionnaire between male and female what should I do?
I am desperately seeking guidance on how to analyze the descriptive stats of a SUBgroup of a larger dataset but I cannot seem to find videos on this, unless I am searching for the wrong title. I have a dataset n=159, but the subgroup I need to analyze is only n=84, and consists of 6 variables within it. Most of my classmates are using some version of Missing Data selections but I just can't figure it out. Can you point me to a video that could help?
What if the categorical variables are divided in class groups like for "Income" Option 1- 0 to 10 $, Option 2- 10 to 25 $, Option 3- 25 to 40 $ and so on. Will this be under category 'Nominal' or 'Ordinal' as dollars ranges are not symmetrical in given options?
That is called "binning" a continuous variable. The resulting variable should be treated as ordinal. You can use cross tabulation with a chi-square to analyze it. Good luck!
Hi, Could you tell what to do if I have 2 habitats where i have 2 groups of rhinos, A-dehorned and B-not dehorned, and i want to test the difference between 2 habitats based on subgroups? Do i do it this way: First for habitat 1 and 2: A & B - normality test- if parametric - paired t- if non parametric - wilcoxon signed rank then the results from both paired tests put through independent test? (data is numerical)
Thanks a lot very helpful! How about if the dependent variable is categorical (the frequencies of nationalities) between two groups (eg: males vs females)? I tried to do it the same ways, but it only accepts continious data.
For that set-up, you would probably use chi-square. The frequencies/counts will be used to weight the nominal variable. I think the techniques are in this video: ruclips.net/video/ysn-YL9bLdo/видео.html. Good luck.
Maybe I can help with non-parametric. This is a summary about them: ruclips.net/video/IwPz5e2y0rM/видео.html Here is a playlist about chi-square: ruclips.net/p/PLVI_iGT5ZuRmr2GuaQU3zDfysDFnoVtxe
How to get the percentages of the Input Data? Such as, for a questionnaire, 40 people agreed, 50 people disagreed and 60 people are Not-Sure to Question number 1. Now, how can i get the percentages of agreed, disagreed and Not-Sure of the given data?
For the single item you described, you have a total of 150 respondents. Divide each response rate by 150 to get a percentage (Agreed = 40/150). Good luck
Thanks for this great video! If I have 80 females and 30 males, may I just compare the means or do I have to consider unequal sample sizes somehow? Thanks a lot!
If you are simply reporting the means, then you can report as is. If you want to do a t-test, you should consider the sample size, but only if your homogeneity of variance assumption is violated. Here is a video about it: ruclips.net/video/OyB_w4XNQ58/видео.html
Hi, I really hope you see this! What happens if one of the groups has no data? For example, If I want to see by how much the exam scores differ between girls in class C and boys in class C, but only the group girls has information (no info whatsoever on boys)? What to do then?
Hi...and yes I finally saw this...if you have no data, your options are very limited. If there is literally zero data, then you can't do any analysis. If you have a count of the number of boys or a mean of their scores, you could report those numbers. I have a video about analysis without raw data...that might give you some ideas. Good luck
No, it does not...but you can make choices during your interpretation of the findings when you have unequal group sizes. For instance, with ANOVA, you might choose a specific post hoc test or you might interpret your t-test slightly differently
@@ResearchByDesign I did a test which revealed the same results as the multiple regression, what post hoc test would I perform? Sorry I'm not very clued up on spss.
This video has greatly helped me in my research. Thank you so much!
I do appreciate if there’s video about missing data from you 🥰🙏🏻 ty
love this!
Very helpful. Thanks, RbD!
My pleasure!
Thank you!
Welcome!
Extremely helpful! You are a life saver!
Glad to hear it!
hi thank you for the video explanation. i want to ask, what if we want to compare a nominal data sample characteristics between 2 groups (nominal vs nominal) ? because in this video only show numerical data (income) vs nominal data (gender). Thank you.
What about correlation between male and females answers? How are we going to do
Hello sir, Thank you so much for this video. But I am still confused I want to compare a lot of questionnaire between male and female what should I do?
V nice efficient and helpful video
Thanks
So nice of you...thanks
Hi Prof, I am confused why AVOVA table is chosen when there is only one independent variable (i.e Gender)?
how to check if the difference of correlation between 2 variables, splitted by gender, is significant ?
I am desperately seeking guidance on how to analyze the descriptive stats of a SUBgroup of a larger dataset but I cannot seem to find videos on this, unless I am searching for the wrong title. I have a dataset n=159, but the subgroup I need to analyze is only n=84, and consists of 6 variables within it. Most of my classmates are using some version of Missing Data selections but I just can't figure it out. Can you point me to a video that could help?
Thank you for creating this! Very helpful for me. :)
I'm so glad!
What if the categorical variables are divided in class groups like for "Income" Option 1- 0 to 10 $, Option 2- 10 to 25 $, Option 3- 25 to 40 $ and so on.
Will this be under category 'Nominal' or 'Ordinal' as dollars ranges are not symmetrical in given options?
That is called "binning" a continuous variable. The resulting variable should be treated as ordinal. You can use cross tabulation with a chi-square to analyze it. Good luck!
Hi, Could you tell what to do if I have 2 habitats where i have 2 groups of rhinos, A-dehorned and B-not dehorned, and i want to test the difference between 2 habitats based on subgroups?
Do i do it this way:
First for habitat 1 and 2:
A & B - normality test- if parametric - paired t- if non parametric - wilcoxon signed rank
then the results from both paired tests put through independent test? (data is numerical)
Thanks a lot very helpful! How about if the dependent variable is categorical (the frequencies of nationalities) between two groups (eg: males vs females)? I tried to do it the same ways, but it only accepts continious data.
For that set-up, you would probably use chi-square. The frequencies/counts will be used to weight the nominal variable. I think the techniques are in this video: ruclips.net/video/ysn-YL9bLdo/видео.html. Good luck.
Hello! Your video was very helpful for me. Can you do the same for non-parametric analyzes?
Maybe I can help with non-parametric. This is a summary about them: ruclips.net/video/IwPz5e2y0rM/видео.html
Here is a playlist about chi-square: ruclips.net/p/PLVI_iGT5ZuRmr2GuaQU3zDfysDFnoVtxe
How can i install SPSS in Macbook Pro???
How to get the percentages of the Input Data? Such as, for a questionnaire, 40 people agreed, 50 people disagreed and 60 people are Not-Sure to Question number 1. Now, how can i get the percentages of agreed, disagreed and Not-Sure of the given data?
For the single item you described, you have a total of 150 respondents. Divide each response rate by 150 to get a percentage (Agreed = 40/150). Good luck
Thanks for this great video! If I have 80 females and 30 males, may I just compare the means or do I have to consider unequal sample sizes somehow? Thanks a lot!
If you are simply reporting the means, then you can report as is. If you want to do a t-test, you should consider the sample size, but only if your homogeneity of variance assumption is violated. Here is a video about it: ruclips.net/video/OyB_w4XNQ58/видео.html
Hi, I really hope you see this! What happens if one of the groups has no data? For example, If I want to see by how much the exam scores differ between girls in class C and boys in class C, but only the group girls has information (no info whatsoever on boys)? What to do then?
Hi...and yes I finally saw this...if you have no data, your options are very limited. If there is literally zero data, then you can't do any analysis. If you have a count of the number of boys or a mean of their scores, you could report those numbers. I have a video about analysis without raw data...that might give you some ideas. Good luck
youre a fucking god
Does spss auto.atically account for unequal grouos when comparing groups?
No, it does not...but you can make choices during your interpretation of the findings when you have unequal group sizes. For instance, with ANOVA, you might choose a specific post hoc test or you might interpret your t-test slightly differently
@@ResearchByDesign I did a test which revealed the same results as the multiple regression, what post hoc test would I perform? Sorry I'm not very clued up on spss.