One-Way Goodness of Fit Chi-Square in SPSS - No Preference (15-5)
HTML-код
- Опубликовано: 26 апр 2017
- We are now going to learn how to calculate a one-way chi-square in SPSS. This is the House of Pigs Chi Square, using the dataset BuildingPermits.sav. A house inspector randomly samples building permits pulled by 70 little pigs who were building houses, to determine whether a pattern exists in the materials that little pigs choose for housing construction. The permits are categorized based upon whether the little pigs apply to build their houses out of straw, sticks, or brick and mortar. Our research question is: “Are the building materials preferred equally?”
This video teaches the following concepts and techniques:
One way chi square with random expected values
IBM SPSS Statistics software
Link to a Google Drive folder with all of the files that I use in the videos including the BuildingPermits.sav dataset. As I add new files, they will appear here, as well.
drive.google.com/drive/folder... 
Хобби
Thank you so much Sir for the excellent video. Before actually using statistics for data analysis in research I could never quite get the hang of it but your videos have not only made using SPSS easy but even statistics fun!
very well explained! thank you sir
I love the formal write up of these examples with "lupus-induced" or "ursine-comfort" before. Good teaching always lighting up the mood and making stats... fun?! :O
Thanks for noticing that. Doing an analysis is just part of your task...you have to know how to write up the results, too.
Big Thanks for this!
Thank you so much sir 😊
I am pretty sure you can double click the hypothesis test summary that you showed at the end and get more detailed information
Yes, that is correct. Double-clicking on the table opens it both for editing and exploration. You can also hover over output (the significance value, for instance) and see more detail, such as more decimal places.
Very good explains
Thanks for liking
Thank you for the great video! I have a question, after running the chi-square and we know there was a difference in the distribution of 3 materials, but we don't know whether it is between A&B, B&C, or A&C. How do we proceed here? Is there something like the post-hoc in ANOVA?
The post hoc is examining the residuals of the groups. Often, you can look at the expected values for the groups and see that one is higher than the other two. Good luck!
Awesome video! How did you find Cramers V?
Hi Prof, Can I still perform Chi-square with two groups of unequal group size (n)?
Yes, but the unequal sample sizes will reduce the power of the test making it harder to find differences All cells must have an expected value of 5 (minimum) or the error rate goes up, as well. Good luck
You really need to adress the result number from the chi test - what do I do with this number 6.5? What does it mean? What can I conclude from getting this result?
When you do a chi-square, you expect that the value will be 1.00, if there is no difference between the groups. The 6.5 tells you that you have some difference between the groups, and the p value tells you if the difference is big enough to be significant. If the p < .05, then the groups are different, or as in this case, the observed values are not occurring randomly.
Nice video but can you tell how u calculated effect size
Try this one. It is about effect size in chi square: ruclips.net/video/sZF7fqNOe7Y/видео.html
I used chi-square and The results of tests seem unreliable as several of my cells have an expected frequency of less than 5 (for each test). The t-test looks inappropriate due to the nature of my data if I am not mistaken. I am wondering how could I compute the mean profitability of two types of farms small and medium sized based on the categorical data that I have. Can you explain how I compute the mean profit of these farms? or use the best appropriate test method? Your reply would be highly appreciated.
You are right that small cells sizes will mess up the chi-square. The best fix is to combine categories so you have fewer cells with more cases in each. As for means, it sounds like you could use an independent t test with "profits in local currency" as your DV and "small vs medium; 1 & 2" as your IV. You would need a scale variable for the DV. If you have a categorical DV and a categorical IV and too few cases in each cell for chi-square, then you need more data. Good luck!
@@ResearchByDesign Thank you so much for your reply! regarding the t-test, if I have a loss which will be a negative number how could I define it in SPSS. because a few numbers of companies have lost. is that enough just to have it negative? a
@@ResearchByDesign I also had same issue where 50% of the cells have an expected value less than 5 :(