turnitin.com

Hi folks,

As if your first finals week at Fuller weren't stressful enough, I heard from several of you last night that turnitin appeared to be glitchy--letting you upload the file, but doing strange things to it or not letting you upload.


The good news is that all files that are currently uploaded to turnin.com (at 6:11 am Friday morning) are correctly formatted. The bad news is that I'm not sure if all the people who emailed me have their files uploaded.


Finally, if you are Elephant Ears, you inadvertently uploaded your output to the writeup file and the writeup to the output file. This will make it impossible for me to use the grading rubric on your final paper, so please correct that right away.


Good luck as you finish up with all remaining requirements, for this class and all your others. Have a blessed and merry Christmas!

Post Hocs in Random Effect Designs

If I ran a random effects design and had no significant main effect (of the random effect variable) and no significant interaction effect, but I did have a significant main effect for the fixed effect variable (which had more than 3 levels), how might I interpret the post hoc? I ran a Bonferroni to determine the differences, but I am noticing it seemed to adjust according the original error term/mean square (the error of the interaction), is this okay to proceed with because it is the fixed factor variable? Or should I be running a different post hoc or interpreting it differently? Thank you!

Also, how might I check homogeneity of covariance? I checked it by splitting the file based on my 2 IVs at once (random and fixed) and then checking the variances of the DV and comparing them to one another, is this correct?

Transforming DV

Hello,

My DV has a slight negative skew. I am only selecting some levels for one of my IVs so I was wondering if I need to fix the skew with the cases I'm looking at or do I fix the skew with all the cases of DV (all 140 participants instead of 118)? The reason I ask is because if I decide to delete outliers the SD will change, which will affect what values are outliers.

My follow up question is for a negative skew what does it mean to reverse the score? And what does it mean to account for change in direction in interpretation?? On the slides it says we can dichotomize, but I thought that is only acceptable for IVs not DVs?

Final project write up

Will we get marked down if we have an explanation of our analysis ( where we also explain any minor violations to our assumptions) seperate than our "Results" section?

Interaction Effect

I know this has been asked a thousand times, but I just want to clarify...is it possible to have no main effects and no interaction effects when running a factorial ANOVA, but then run a post hoc and find an interaction effect? Or I am I done once the original print out shows there is no main or interaction effects?

Retake Exam

Do we know where the retake exam is going to be held yet?

Final Project

I am still confused, the book says that if you have a negative skewed distribution the best thing to do is to " reflect the variable and then apply the appropriate transformation for positive skewness" pg 94 last paragraph. is this correct or can we just reverse the score then square root?

Interaction effect

Is it necessary to predict an interaction between two variables or is it okay if you pick two that you think will have main effects, but no reason to have an interaction effect?

Thanks!

Alternate Exam

I emailed Dr. Clements to ask about how grading will work for the exam. She said that if we take the alternate exam, she will take the highest of our two test grades! :)

-Meredith

ANCOVA covariate outliers

If we run an ANCOVA and the CV has outliers, should we recode the outliers of the CV just as we would for outliers on the DV of a regular ANOVA, or would this be something not permitted and simply having outliers on the CV at all would mean it shouldnt be used?

In regards to the violations of the assumptions of ANOVA, do I write that in the write up and indicate how I dealt with it? Also how specifically should I explain my treatment of the violations.

At trying to change the skew of my data, do I do all reverse, square root, logarithm, and inverse? or can I just go as a far a I need in order to reach a normal distribution?

Also what is considered to be a bad skew, is a skew beyond negative or positive 2.5 a bad skew?

Final project

Do we still report on descriptive statistics, even if we have non-significant results?

Final project

I have a negatively skewed distribution and i would like to correct for this by reversing the scores...How do i do this on SPSS?

I also have unequal cell sizes, but my Levene's test is not significant should i still f/up with an fmax of <3 ?

final project

Do we have to get significant results for the main effects and interaction effects in the final project. Or, is it okay if that there are no significant findings?

Rescheduled Exam

I have reviewed the responses to the survey, and read email suggestions about how to fairly address the exam. The following plan is in place.

The Tuesday/Thursday section (and only the Tuesday/Thursday section) will take the exam on Tuesday at 8 a.m. in Room 116.

On Wednesday at 1 p.m., all students from either section who would like to retake an alternate form of the exam will be permitted to do so. The location of this make up exam will be announced later this week.

All three administrations of the exam will be curved separately (assuming that a curve proves necessary). That is, the administration from last Wednesday will have its own curve, the administration from this coming Tuesday will have its own curve, and the administration from this coming Wednesday will have its own curve.

I am aware that not all students are able to take the exam at these times. If you are one of those people, please email me directly.

final project

i recoded variables to dichotomous variables and since I have, I no longer get any values for the interaction between variables- the main effects are there but interaction effects are blank. Can someone please let me know if this is okay/normal or what I could have done wrong and how to fix this? Is there another way that I can test for interaction effects?

Retaking Test

I think that the T/TH class should be able to see the exam that the M/W class took if they are going to be allowed to retake the exam. Even though the questions will be different, they would have an unfair advantage since they have been exposed to a test on this material already. I think that would be the most equitable situation.

Retaking Test

Final project

after we choose our variables, do we need to run ALL the analyses that are possible with the variables we chose?

will we get marked down if we only run one analysis ?

retaking test

Dr. Clements,
I was wondering if M/W class will have the chance to retake the test along with T/Thurs? Does that mean that there will be some sort of curve since M/W had the chance to look at the format of the test already?

In regards to the final project, can I delete my outliers and inverse my scores to fix the skewness. Even though I will still end up with a slight kurtosis if I do that?

For the final project, is it okay if I choose certain levels of the IV to work with?

Exam 3 Rescheduling Survey

As you know, Exam 3 must be rescheduled for the Tuesday/Thursday section. This raises interesting issues on both pragmatic and fairness fronts. Obviously, the exam must be rescheduled for a time and date that works for those enrolled in the Tuesday/Thursday section. Because it was unclear whether or not the campus would be open today, and today might not have worked for all students, it was not possible to schedule the exam for today. This raises the issue of greater study time for the Tuesday/Thursday section as compared to the Monday/Wednesday section. To that end, I am asking that you all (i.e., Monday/Wednesday students and Tuesday/Thursday students) take a survey about Exam 3: Click here to take survey Please respond by Saturday evening if at all possible.

In my DV, I run a Q-Q plot and a histogram, it looks fine, but how can I ensure that it is not skewed. If I compute my z score, could the z score be negative or does it only have to be positive?

IMPORTANT: Reporting Non-significant Results: Addendum

Okay, so in a previous blog, I said that you don't need to report the inferentials on non-significant results.

This is generally true, BUT, if your hypothesis PREDICTS significance and yet the FINDING is non-significant, you should still report the inferentials.

You can look on page 32 in the APA handbook for the full discussion on the results section, but the relevant portion is this: "...be sure to include small effect sizes (or non-significant findings) when the theory predicts large effect sizes (or significant ones)."

Why? Well, most hypotheses are generated on the basis of past research. So if your hypothesis comes up short, than that (may) actually cast doubt on previous research too. So only fair to report your inferentials because you are actually reporting something new: that perhaps what was thought to be a relationship between variables is in fact not a relationship.

In short: NON-SIGNIFICANT results (statistically speaking) can still be significant (generally speaking) if they run counter to either past research or your current hypothesis. So report your inferentials for these sorts of cases of non-significance.

This could be important for your final project, because it is very possible that your hypothesis may not generate a statistically significant finding, and you will still need to report the inferentials.

Data Transformation

If we transform our dependent variable due to skew and outliers, where do we report these transformations (so the mean values reported make sense to the person evaluating our final project)? Do we just state them in the section of the project where we are explaining the analyses we used and why?

Fmax

While working on the final project, I violated the assumption of homogeneity. My cell sizes are discrepant by a ratio of 5:1. Since it is more discrepant than the 4:1 ratio, I know my Fmax value needs to be lower than 10 but since it is not quite to the 9:1 ratio, do I have to lower the critical value to less than or equal to 3? My Fmax value is 3.6 and I am thinking that ANOVA should be robust to this violation since my cell sizes are not necessarily close to the 9:1 ratio discrepancy, but I just wanted to check.

Campus is Closed; Exam to be Rescheduled

As you likely already know, the campus is closed due to Santa Ana winds. The test will need to be rescheduled.