This series about writing results is very helpful! Thank you! It would be interesting to see a video with proper examples of the analysis of Likert scales and nominal variables in questionnaires.
I just love how detailed you explained on how to report the result in APA style, and I love how you make it so fun, I love how you crack yourself up, it brings smile to my face.
Great vid! How much description/interpretation are you allowed to do before it becomes information that should be in the discussion? If I were, say, comparing effect sizes of significant predictors, is that okay to still be in the results?
Yes, you can add that. This tutorial is geared toward undergraduate students, where it’s more about reporting the numbers and then leaving the interpretation to the discussion, but in practice, there is a great amount of flexibility in what goes where. If it is numbers-based, Results is fine for comparison
Would you report the semi partial correlation in the values section of a written coefficient results section? i.e., Motivation was a strong, positive predictor of academic performance accounting for approximately 14% of the unique variance in overall first year grade, β = .405, sr2 = .143, t(115) = 5.937, p < .001.
Thanks! If I wanted to look at the effect of IV on DV whilst controlling for variables such as age/height etc, would it be the same thing? Or would it have to be reported differently?
MR is for continuous variables, typically. If you have a categorical IV, you may want to do an ANCOVA. It’s effectively the same thing, math wise, so it just depends on your variables and how you set them up
Thank you for the great video! It's really help me with my assignment. However, I'm curious how you would report two significant variables? For example, if both dan and the baby's sleep were significant. Do you report them separate as well as give separate explanations? Or are they reported together?
Great question! Yes, report them both, briefly explain them both, separately, but within the same paragraph, since you are reporting all the coefficients of the same model.
Thank you!@@AlexanderSwan that helps a lot. If it's okay, I want to add one more question? If I were to then add more variables (that I consider to be nuisance variables) and wanted to check if the current model's result is independent of these new variables, what kind of regression would I be looking at? My head says hierarchical of some sort but I can't seem to find any definitive answers...
I love this Video. But isnt say something like "This means that Dans overall level of grumpiness decreases by nearly nine points on the Scale for every hour that he gehts sleep" implying causality, where might be no causality?
I would say it is a statement of fact within the result section. However, if you continue to use the same language in the Discussion, then you are trying to make a causal claim. So it depends, I’d say, where this statement lives.
You have an upbeat and engaging delivery of statistics! Such a breath of fresh air! Sincerely, thank-you :)
This series about writing results is very helpful! Thank you!
It would be interesting to see a video with proper examples of the analysis of Likert scales and nominal variables in questionnaires.
Oh my gosh !! Wow thanks so much those details are exactly what i needed clarified for my research project! thanks so much - so helpful!
I just love how detailed you explained on how to report the result in APA style, and I love how you make it so fun, I love how you crack yourself up, it brings smile to my face.
Great vid! How much description/interpretation are you allowed to do before it becomes information that should be in the discussion? If I were, say, comparing effect sizes of significant predictors, is that okay to still be in the results?
Yes, you can add that. This tutorial is geared toward undergraduate students, where it’s more about reporting the numbers and then leaving the interpretation to the discussion, but in practice, there is a great amount of flexibility in what goes where. If it is numbers-based, Results is fine for comparison
Would you report the semi partial correlation in the values section of a written coefficient results section?
i.e., Motivation was a strong, positive predictor of academic performance accounting for approximately 14% of the unique variance in overall first year grade, β = .405, sr2 = .143, t(115) = 5.937, p < .001.
Yes, that is fine. For these additional stats, include where they logically fit in the narrative, overall.
Please make a video on Stepwise multiple regression analysis and how to make the final table.
This is a good idea - I’ll see what I can do soon
@@AlexanderSwan thank you, i will wait for the video.
Thanks! If I wanted to look at the effect of IV on DV whilst controlling for variables such as age/height etc, would it be the same thing? Or would it have to be reported differently?
MR is for continuous variables, typically. If you have a categorical IV, you may want to do an ANCOVA. It’s effectively the same thing, math wise, so it just depends on your variables and how you set them up
Thank you for the great video! It's really help me with my assignment. However, I'm curious how you would report two significant variables? For example, if both dan and the baby's sleep were significant. Do you report them separate as well as give separate explanations? Or are they reported together?
Great question! Yes, report them both, briefly explain them both, separately, but within the same paragraph, since you are reporting all the coefficients of the same model.
Thank you!@@AlexanderSwan that helps a lot. If it's okay, I want to add one more question? If I were to then add more variables (that I consider to be nuisance variables) and wanted to check if the current model's result is independent of these new variables, what kind of regression would I be looking at? My head says hierarchical of some sort but I can't seem to find any definitive answers...
@@charlottecoetzee8298 yeah, hierarchical!
can you do this for a multiple logistic regression?
A good question, but it’s not quite the same, unfortunately
do you look at adjusted Rsquared or multiple Rsquared
It depends on if you need to bias-correct. Most of the time, you just look at R^2 and report that.
Love your videos! Thanks!
I love this guy
Aww I love you too
Wish we could see how you make the APA tables
I have a video on this. Check out my channel
I love this Video. But isnt say something like "This means that Dans overall level of grumpiness decreases by nearly nine points on the Scale for every hour that he gehts sleep" implying causality, where might be no causality?
I would say it is a statement of fact within the result section. However, if you continue to use the same language in the Discussion, then you are trying to make a causal claim. So it depends, I’d say, where this statement lives.
@@AlexanderSwan Thank you for taking the time to reply :)
Next time please explain slowly
You can use the player controls to slow down my voice/playback. You can also play/pause the video as much as you need!
In other words: don’t treat it like a traditional lecture!
@@AlexanderSwan alright. Thank you
@@AlexanderSwan got it