Final Project Dataset

I was wondering when the dataset 851_final_project.sav will be posted on Portico? Also, is it under a different label somewhere on the ANOVA cd we have? Thanks!

I am having problems understanding what is orthogonality, its relationship to sums of squares (thus it impact in the rest of the inferential statistics) and the benefit of orthogonal factors.

I don't understand either how the degrees of freedom are computed in Repeated Measures(seems to be K-1, N-1).

Finally, when you say in the review guide "Multiple Comparison Procedures", do you mean the epsilon adjustments of df used for correction of violation of sphericity?

Article Summary

I have a study that did not specify ethnicity, so I did not put that in the sample, but I was marked of because sample was not described. I was wondering what should I do if the study did not provide any information regarding that? And also, what are the things that generally need to be reported, whether or not the information is given or not in the study?

ANCOVA Question

At the review session on Tuesday, the question was brought up as to how you tell whether or not a covariate is related to an independent variable (a relationship you do not want). At the time, I said I thought it was done by running a simple correlation. As I was looking over the ANCOVA slides today, though, I discovered I was mistaken.

Let it be known that in order to test if there is a systematic relationship between the CV and the IV, you can run an ANOVA with the CV as the DV and the IV as the IV.

This should especially be done with nonexperimental designs.

Sorry for my mistake, and I hope this clears things up!

As a reminder...

The weekly review sessions have been scheduled to allow you to consult with the TAs on questions you have. Some of you have requested, through me, that you have an additional session with the TAs, and those requests have been approved. Please, however, keep in mind these things:
1. An additional meetings scheduled with the TAs are ADDITIONAL meetings that can only be scheduled if you are, in fact, attending the original meetings (i.e., the review sessions).
2. Additional meetings should be cleared with me rather than approaching any of the TAs directly.
3. Questions that do not require a meeting should be posted on the blog (not on Facebook, not through emailing/calling/contacting the TA directly). As we discussed earlier this year, it is imperative that I know what questions you have (i.e., what is unclear), it is imperative that I know how your questions have been answered (i.e., what aspect of the question was attended to and how), and it is imperative that we do not use up the limited hours available to each TA by having different ones of them answering the same questions. Remember that if you have a question, it is likely that someone else also has that question, so consider it your civic duty to share your questions.
4. Remember that your TAs are also dealing with their own courseloads, clinical work, and other responsibilities, and that their resources are finite.

data set for ANCOVA homework

Just to be ultra careful, the dataset we use for the ANCOVA homework is again, "repeated measures and ancova.sav"?

Or am I wrong?

Thank you.

Blessings,
Ramona

review for Exam 2?

I was unable to make it to the review session today due to Live Team...will there be another review session for exam 2 this week?

Standard Error

A question came up at the review session about standard error, specifically if it can exceed one. If it can, you would need a zero in the ones place should there not be another number there.

After looking around and consulting with a fellow TA (thanks Ann!), let it be known that it is possible for standard error to exceed one.

One cautionary note, though: just because it is possible for standard error to exceed one, does not mean that it is possible for every variable's standard error to exceed one. Know what your variable is before deciding whether or not it can exceed one. (Think back to your factorial homework where the DV was body fat index, which was a proportion, which could not exceed one. Therefore, your standard deviation could not exceed one, and you should not have had a zero in the one's place.)

Estimated Marginal Mean

I only get 1 mean for this one, and I thought I was supposed to get 2, 1 for PREP and 1 for No PREP? And the Grand Mean that was given to me also had an "E" in it. Is that how it's supposed to be?

Problem with Estimated Mean

After running the analysis, the estimated marginal means in the output came out with an E (eg. 1.234E5). What happened?

Article Reviews 5 & 6

Article Reviews 5 & 6 are at the front desk except for the five for which I do not have names. As before, it is to your advantage to pick up your article reviews and to carefully attend to what is marked off on the grading sheet.

I need pseudonyms for Sun Choi, Vanessa Klesert, Rachel Langford, Ryan Poling, and Garrett Woods. (This is the one for which I did not have a pseudonym sheet, so I asked you to come up with three random words.)

Exam 1

There were a few questions on exam 1 that I missed that I would like to double check again to make sure I understand - Would it be possible for me to look at it again?

Formatting of responses

For all homework assignments and the final project, you have been provided with coversheets on Portico. Most of you are attaching these coversheets, as instructed, which is good. Most of you are not, however, answering the open ended questions in the spaces provided on the coversheet.

Please don't re-type the whole coversheet. That's a waste of your time, and it does not serve the purpose of the coversheet (which is to make it easy to find the answer to the questions by having them in the same place).

Please confine your answers to the space given on the coversheet. If you are concerned that the TAs or I can recognize your writing, please don't be. We can't (and we don't even try to do so.) If you are concerned that we cannot read your writing, please make the effort to write neatly. If you absolutely must type your response, literally cut and paste/tape it onto the cover sheet. It is all too easy to miss a response that is inserted between printout pages, and neither the TAs nor I want to fail to give you credit for an answer you really knew.

Review Session Questions

A few questions from the review session were raised in class today, and I just wanted to clarify these items.

1. Multicollinearity is just as you learned in regression. It refers to predictors in a regression being too highly correlated with each other. In ANCOVA, the regression that is computed is the regression with the dependent variable being predicted from the covariates. Thus, multicollinearity can only occur in ANCOVA when the covariates are too highly related to each other. The IV and the DV are both irrelevant for the examination of multicollinearity.

2. Having said that, it is also true that the covariate should be unrelated to the IV. This, however, is not multicollinearity, but independence of covariate and IV.

3. You should report a measure of central tendency (i.e., the mean), a measure of variance around that mean (i.e., SD or SE), and an n for the cells you actually analyzed. This means that in the second homework problem, you should have reported the means, SD, and n for the cells defined as PREP at Time 1, PREP at Time 2, Control at Time 1, and Control at Time 2. You should not have reported the marginal means for PREP (averaging over time) or Control (averaging over time) or for the total sample (averaging over both group and time) or for time 1 (averaging over group) or for time 2 (averaging over group): none of those are the model you actually analyzed.

Finals

I was looking at my final exams schedule and I realized that there are 3 exams scheduled on Tuesday during final week. This will apply to anyone who are taking the tue./thur. ANOVA, psychometirc and OT writings this quarter. I don't think I can handle that many exams on the same day, can we reschedule our ANOVA final on Monday (taking with the Monday/Wednesday class) or any other day during final week? Thanks.

Reporting the Means...

Thanks for the helpful information regarding the means we have to report. But I still have some questions about how to report them in our results section correctly.

1. For repeated measures design, we only report the N once. Do we simply report the N with other descriptive statistics (M and SD) of the first level of the IV we put in the results section and report only M and SD for the second level of the IV?

2. For mixed designs, since we have to report all the means for different levels of IV, can we put descriptive statistics of the randomized IV in a table and descriptive statistics of the repeated measure IV in text? Or, do we have to put all the descriptive statistics in one table? If the correct way of doing it is the latter case, how do we fit in both ns and N in one table?

Thank you!

ANCOVA File

Is anyone else having trouble opening the repeated measures/ANCOVA file? I've been trying for about 15 minutes and all I get is some sort of text conversion wizard in SPSS.

Because Mulitple People Have Asked...

In a Repeated Measures Design, as with any ANOVA design, you must report the means, the standard deviations, and the ns for the cells you actually analyzed.

The reason for this is because ANOVA apportions variance over all variables included in the analysis. Had you not included time in the design, the obtained F for newgrp would not be the same as the one you got for the homework.

That is why including only the marginal means was incorrect on both your last homework and on the exam, and would be incorrect on this homework as well.

My question is regarding the second step of the assigment. We are suppose to include intervention status as a between subject factor. When I compared this assigment with the example discussed in the classroom I noticed that when we examined gender and time we included both in the box for within-subject factor and then we examined as many groups as combinations were possible with this two variables (e.g. pre-females, pre-males, post-females, post-males, etc.). Putting the newgrp under between subject factor do NOT allow us to have as many groups as possible combinations of the variable (Prep Time 1, Prep Time 2, No-Prep Time 1, No-Prep Time 2). It also happens that just Time is analized in the Mauchly's Test of Sphericity and in Tests of Within-Subjects Contrasts the group intervention is not analyzed, just Time and the Interaction between Time and newgrp is included. Is that how it is supposed to be? If yes, why is different from the example discussed in class? Do we need to create the groups as computing a new variable?

Repeated Measures

Q2c:

for the within-subjects group, we were measuring males, but later on when the intervention was added, the group was made up of both genders (couples); how does that work?

Q regarding Repeated Measures Assingment

I wonder how to interpret the significant effect of between-subject factor, (which is "group" in the second analysis in our repeated measure assignmet) in a mixed design model. In addition, what further tests do I need to do with this significant result?

Repeated Measures Homework newgrp label

Hi, I converted the repeated measures homework from excel file to spss sav. file, so, this question may only have to deal with those who did the same thing. I converted it succesfully, but I found that the newgrp variable only has 1 and 3 labeling the groups instead of what Dr. Clements has told us that there will be 1 and 2. So, I assume 1 = PREP, and 3 = control. I just want to make sure whether my assumption is right or not. Thanks!

Exam 2 Review Sheet Updated

As of 3:45 today, there is a new review sheet for exam 2 posted on Portico.

Hope your other exam went well...

Monday Wednesday Syllabus Update

To reflect both the error in due date pointed out in class yesterday AND the efficiency with which Justin presented Mixed Designs yesterday, the Monday Wednesday syllabus has been updated and is posted on Portico.

Factorial Homeworks at Front Desk

Please pick them up.

From the Writing Center

Please encourage any or all students to attend the upcoming Writing Workshops, provided by Fuller's Writing Center. We hope that these workshops will benefit both students and faculty by boosting the quality of student writing and thereby lowering at least one common source of faculty frustration! The schedule for this quarter (also listed in this week's SEMI) is as follows:


Composition Workshop
How to develop a strong thesis statement, form coherent paragraphs, and use supporting evidence appropriately
Friday, February 13, 1-3 pm

Advanced Composition Workshop
How to develop different kinds of thesis statements, transition deftly between paragraphs, smoothly integrate quoted material, and more!
Friday, February 20, 1-3 pm

Grammar & Editing Workshop
How to fix common grammatical errors such as inconsistent verb tense, flawed sentence structure, and misplaced punctuation
Friday, February 27, 1-3 pm


All workshops will be held in Payton 102 and are FREE (with an optional resource packet available for $5). Workshops will be presented by Kathryn Streeter, a recent graduate of Fuller's MDiv program and an experienced SOT TA. Feel free to contact her with any questions at kathryn_streeter@fuller.edu.

test reviews coming up

hey friends,

just a heads up, since exam 2 are on a tues (2/24) and wed (2/25) instead of the normal wed/thurs, the test reviews are slotted for wed (2/18) to prep for the tuesday test, and tuesday (2/24) for those who might be taking the test on wed that didnt make it to the earlier review. there will still be the normal weekly review session on tues (2/17).

just wanted to give you guys a heads up in the midst of all your assignments and tests coming up.

Repeated Measures Homework Dataset

It's very possible that I could be missing something here, but I'm having trouble locating the SPSS data file for the repeated measures homework. The assignment states, 

"Using the dataset repeated measures and ancova.sav, examine the effect of time on dissatisfaction with sexuality in a sample of newly married men."

However, I can't find a data file with either of these names on the T&F cd or Portico under the files section.

Thanks for some clarifcation on this issue....

Chris

Picking Up Homeworks and Article Reviews

I needed to use Katy's computer to enter test grades, and while I was at the front desk, I also went ahead and filed the remaining graded article reviews. I got both a surprise and an answer when filing them.

The surprise was that there were assignments from last quarter that folks had not yet picked up.

The answer was to my question of why things that have been corrected on previous assignments continue to be missed.

PLEASE pick up your assignments. The most recent article reviews reflect the correction that one of you actually pointed out about capitalization of headings. Congrats to that student for the careful reading of the APA Publication Manual.

Please also stop by the front desk to review your exams. As before, you will not be able to take them, but if you have concerns about your test, please sign up for a time to meet on my office door, and I will be happy to go over the exam with you. (FYI--the mean exam grade was 8.6, and this was also the median grade. The modal grade, however, was 100.)

Unequal Cell Sizes

After reading pages 216-220 in the textbook on unequal cell sizes, I am still having a hard time comprehending how to tackle question 2F. Is there anywhere in our notes or in the book that more clearly addresses the issues of unequal cell sizes in factorial ANOVA? Also, is nonorthogonality an issue for an interaction variable in a 2 x 2 factorial design, or is this only the case for the highest interaction term in higher order designs? Thanks!

Factorial Homework Question

Do we have to report the partial eta square, F ratio, and the probability level of non-significant interaction effect and main effect? If we have to report them, do we report them in-text or in a table? Thanks.

ANOVA CD found in the library...

hey guys,

my roommate found an ANOVA CD laying around in the 3rd floor compy lab. if you are missing yours, let me know and i will get it back to you.

hope week 5 is treating you well!

Reporting Factorial ANOVA Descriptives

If a factorial ANOVA has a non-significant interaction, is it necessary to report any descriptives besides group differences on levels of one IV that are averaged over other IVs (e.g., only report marginal means, associated SDs and cell sizes, rather than factorial cell means, SDs, and cell sizes)? It seems to me the only time it makes sense to report indivudal group means (factorial cells) is when there is a significant interaction...

Thanks

Chris