Sum of Squares Types

Also, in determining type, how are we to know whether the data sets we were given were "designed" to have equal cell sizes or not?

Factorial ANOVA Practice

For the Factorial ANOVA Practice, I have two questions:

1.) I am trying to figure out which post hoc comparison to run. In the chart on page 133 of our book, it asks for the type of comparison... complex or pairwise. What is the difference again? From here, if we are doing a complex and decide we should use a modified Fs, how do we select this with SPSS? I know it means modified Scheffe, but SPSS does not list that option. If we were to go with Pairwise and choose Tukey-Kramer, I do not see this option either in SPSS. What do you recommend?

2.) If we have unequal cell sizes, is there a way to specify to SPSS to use Type I or Type II Sum of Squares?

Thank you for your help!

Brooke

Creating new variables for post hoc?

If there is no significant interaction, do we have to create new variables for the post hoc?

One more question!

For #8, was the assumption of normality violated? My skew(upon visual inspection of histogram graph) was negative, although every other rule for normality was correctly positioned with our data. All our notes say is "inspect the graph with large samples". The scale used (life satisfaction) was bound to result in skew ("are you satisfied with life: rate 1-3"?!?) but that's an issue with data set up, not analysis. Furthermore, should we be adjusting for skew with our data (transforming for negative skew according to our notes)? Because that sounds like an awfully big adventure I might not be emotionally prepared for...

More ANOVA

In the sample factorial write up, the graphs were included in the write up. In our own write up, should we do the same and say "See graph" and include those graphs (according to APA formatting, of course) that are relevant?

:)

ANOVA Homework

I noticed in the factorial homework that in the case of the life quality univariate analysis that while we had a significant ANOVA result regarding region (p=.036), there was no significant result in the post-hoc Bonferroni analysis in any of the 3 comparisons. Is this possible or is my analysis faulty?

Thanks!

selecting cases

Hello. Some of us had a question regarding selecting cases in the SPSS.

In the practice hw of the Oneway ANOVA, we were told to put in the command LT 3 to get rid of the Independent variable's group 3. That works. But in the case that we want to get rid of the second group of the IV instead of 3, the command LT 2 does not seem to work. When the LT 2 command is given in the selecting cases, it gets rid of both the second and third groups of the IV. The command LT 1 gets rid of all cases of IV. So I guess we are wondering if there is a different command that we can give instead of LT? The only thing that seems to work is LT 3 to get rid of the 3rd group in the IV.

Thanks.

problem opening spss factorial anova

Either time for the review session would work for me. Also, I can't seem to open the Factorial ANOVA homework data set on SPSS.

Weekly Q&A Session

Hi all--

The TAs and I were contemplating putting into place a weekly question and answer hour to supplement exam review sessions. Unlike the review sessions, the TAs would not have any sort of presentation to give or set topics to review. Instead, this would be a time when members of the class could bring their questions, and the TAs would respond.

From last time's scheduling, we know that Tuesday at noon and Friday at noon are both generally workable times. However, we don't know if the greater attendance Tuesday at noon was due to the exam being Wednesday or to that being a better time.

If you think you might take advantage of such a Q&A time, comment on this post with whether Tuesday or Friday is better.

Thanks!

Spaces after periods

Hey guys!

You are doing very well on the article reviews. We just wanted to point out that with the APA manual, 6th ed., you are technically supposed to have two spaces after the period when ending a sentence (see p. 88).

Wow

Lots of questions while I was on an airplane coming home from the conference.

I just want to take this opportunity to thank you all for making good use of the blog and to thank the TAs for their excellent answers. Great job, everybody!

Strange APA style question

So I know spacing is important when we report statistics, so if we are reporting a negative number and the number falls at the end of a line such that the negative sign stays on the line while the rest of the number goes on to the next line, is it appropriate to add additional spaces so the negative sign stays with the number?  (I know its a weird question, I just want to do it correctly)

Example:

There was a significant difference between the groups, t(140) = -
3.4, p = .032.

ANOVA HW 1 Q. 2

Will we be docked points for leaving in outliers if we specify that we left them in in our results section?

ANOVA HW1

I am really confused on the normality assessment of our homework. I believe that there is at least one outlier, and I have tried using the ways in the book and in our notes to determine for sure, but am missing one piece of the puzzle with them all. I cannot figure out with SPSS how to get the box plot or the histogram to work. I tried using skew and kurtosis with z score formulas, but I can't figure out on SPSS how to get it to show me my skew or kurtosis score, and I don't know how to figure out the standard deviation for the population (s (sub N-1)). Any help would be greatly appreciated.

OUTPUT Questions

Hi!

I am having a really tricky time exporting my SPSS output into a word document; however, it is exporting really nicely into a pdf. file. Is it okay to submit my printout that way? Thanks so much for your help! Cheers!

Designation of levels of IV

Aside from the APA style results sections, can we just specify the different IV levels as level 1 or level 2 or level 3 of IV or should we continue to call them by our designated names ex. in my case, 10th grade, 11th grade or 12th grade?

Calculating Fmax

The Fmax formula is the largest variance/smallest variance.

Is this for the entire level of IV or for separate cell comparisons?

For example, as in the hw, with three levels of IV, will there be one Fmax (largest variance out of the three levels/smallest variance out of the three levels)?

Or three Fmax depending on the levels of comparison? Ex. In comparing Level 1 and 2, if the variance of level 2 is bigger than 1, then the fmax will be variance of 2/variance of 1.

Degrees of Freedom

I can't remember... how many degrees of freedom do you round to? Zero, one, or two decimals? I don't believe it's in our notes.

Fmax

I am a little confused on the fmax test... if it is < 10, do we use equal variances assumed or not?

If we do a data transformation, like a square root or logarithm of the data due to skew, the transformations should be run first, and then we should run the t tests and the ANOVA using the data from the transformations rather than from the original data, right? Thanks so much for your help!!

Homework question again

I am having the same problem getting the one way ANOVA to only test two groups like the last question of the homework asks. Does anyone know how to get ANOVA to only test two groups? Every time I try it tests all three levels of Variable 1 instead of only two of them.

ANOVA Homework

I am having trouble running multiple t-tests of the three levels of Variable 1 on the homework. I can't figure out a way for SPSS to divide Variable 1 into three groups that I can run individual t-tests with. Every time I try I can only compare the mean of variable 1 with the mean of variable 2. How can I make it SPSS run the amount of individual t-tests that I need it to run for this assignment?

Copying and Pasting into Word

I am trying to copy my outputs into word, but the only thing that will paste are my Histogram and Q-Q plot. I copied and pasted everything before, but now it won't let me. Any suggestions?

ANOVA HOMEWORK #1

Are the SPSS printouts for the homework supposed to be the exact outputs from the SPSS program? Or do we have to create our own tables with the calculated values? I am just a touch confused about this part of the homework assignment...Thanks so much for your help!

Questions on OneWay ANOVA Hmwrk

Ashli asked:

For the OneWay ANOVA Homework, would you like us to answer the questions and write our APA style paragraphs within the assignment document? Or would you prefer for us to answer all the questions about all the types of tests in one results section as if it were part of a single APA article? Or would you like us to have separate paragraphs as if each were its own separate results section in APA format?

Also for the violation of normality on the homework do we need to go back and recode the 4-6 outliers or since it passed the Fmax is it okay?

One Way Homework

What is the best way to determine which post-hoc test to use?

Good luck on the exam!

May you remember everything you have learned!

Question Re. Test

The practice homework also asks us to run a post-hoc test, and which one would we use... we have not gone over this in class, so will we need to know this for the exam today?

Brooke

Article Review Grades

I was wondering if there is a way to know specifically what we lost points on in the article reviews (v. what are just comments to improve our writing/formatting/etc)?  Or is the grade is more subjective? 

Thanks.

Questions on the Practice Homework

1.) I cannot seem to find how to report results as text in the APA publication manual. Does anyone know the page?

2.) On the practice grading rubric, it says "direction of effect reported" for both the t-test and one-way anova. How/ where do you find this?

3.) E. asks "Do you have outliers?" ... is there a way to tell without graphing?

4.) G. asks "Have you met the assumption of independence of observations? What information is critical to you in knowing if you have or have not?" ... how can you tell? By the effect size?

Thank you!

Brooke

3 Questions

1.) On the Review sheet, #3, it tells us to know which variable types are appropriate as IVs and which are appropriate as DVs in ANOVA. I know we went over variable types in class and have them on our slides; however, I cannot find which are appropriate as IVs and which as DVs. In the book, it simply says that IVs include both qualitative and quantitative, and DVs, continuous, ordinal, and discrete. Any further clarification would be greatly appreciated.

2.) 1st Handout, page 4, slide "So You Have Heterogeneity of Variance", Understand Why... 1. Why is it that the spread of scores is different in different groups... is this because of the IV or ?? and 2. Is it that the level of IV affects participants differently??

3.) At a review session I wrote in my notes, that "a smaller alpha is less conservative"; however, a greater alpha would increase a type I error, would it not? I would think it would think that a smaller alpha should be more conservative? Clarification would be greatly appreciated.

Thank you for your help!

Brooke Yetter

2 more questions on the 2nd class handout

1. In the 2nd class handout, page 2, the first slide, the Levene's test came out significant so I looked to the bottom column to find the t-value. In reporting this t-value though, the degree of freedom is not a whole number. Is there any rule for reporting df values that has decimal points?

- Can you reiterate how to report the t-values again?

2. In the 2nd handout, page 3, the third slide, Dr. Clemente did a post-hoc test for the little league groups and used a Dunnett T3. However, the data did not violate assumptions of homogeneity and the N = 191, clearly bigger than N<75. Can you explain why she used the Dunnett post-hoc instead of the others in this case?

Thanks!

partial eta squared

There was a question about the difference between eta squared and partial eta squared. For a one way ANOVA, they are the same because SS(tot) = SS(effect) + SS(error). However, as analyses get more complicated, this won't always be the case. In these cases, partial eta squared involves partialling out (removing) the other SS terms besides SS(effect).

If all that was confusing, just know that in this class, we will mostly only be concerned with partial eta squared.

Review Article Formatting

I was just trying to turn in my article reviews on turnitin.com, and I uploaded the file from my computer. However, when it showed me the preview of my file, all my spacing and formatting was messed up, including the reference. Is there a way to fix that, or does the formatting get changed back somehow when someone goes to grade it?

REVIEW SESSION

hey guys!

Just wanted to remind you that we will have a review session for Exam 1 tomorrow from 12-1. We will be in Room 126.

2 questions:

1. One of the articles I'm reviewing involves hierarchical linear regression, and I am rusty on regression anyway, and I am confused about how to interpret those statistics-- the chart shows statistics for Beta and R^2 change for several variables. What should I report... the r^2 change?

2. I am still confused about using the issue number of a volume in an APA citation... can you explain when you would put the issue number in the citation and when you wouldn't?

Oneway ANOVA hw

In the 1st hw, the question

"How did you examine whether or not the assumption of normality was violated in your data? Was it necessary to address a violation? Why or why not?" appears twice.

I suppose the second question should read "homogeneity of variance" instead of "assumption of normality" correct?

One Way Data set

I am having trouble downloading/viewing/accessing the onewaydataset.sav. Do I need a particular program to access this?

Thanks!

Article Review #1

All Article Review #1 assignments have been graded. Please carefully review the comments and scores on turnitin.com, and make sure you understand where you lost points so that you may do even better on subsequent article reviews. (If you do not yet know how to see the comment in turnitin.com or how to access the grading rubric, please use the tutorial/help features on turnitin--the TAs and I want you to benefit from the feedback we have left you.)

If your assignment was submitted late, the posted score does not adjust for that, but the corrected grade kept in my database does.

Article Review Grades

1. If we have a very small sample population and no information about demographics/backgrounds of participants studied, are there specific things we can be looking for in the limitations/flaws of the study?

2. How many points are the article reviews graded out of?

Review Session Questions

1) Is Levene's test only looked at if cell size ratios are greater than 4:1, or is it always used?
Levene's is always looked at as a preliminary step. If it's significant, always follow up with an Fmax test.

2) Which row for Levene's test should be used: equal variances assumed or not assumed?
I think there is some confusion here. There's only one line in the Levene's test. The equal variances assumed and equal variances not assumed lines are for the t values in a t test, and the Levene's test tells you which of those lines to examine.

3) What is the APA standard for reporting a p value of .0498? p < .05? p = .05?
The only time < is used is when p < .001. This would be an unfortunate p value to obtain, and if one did obtain it, one might consider whether or not putting the values in a table were possible.

4) Should they know a quantitative way of determining outliers? (p. 89 of the text includes a general rule of thumb based on z)
Not necessary. We will get to box & whisker plots, which are probably the best graphic way to identify outliers, but for now, just think about what an outlier might look like on a histogram or a Q-Q plot. (But great that you are reading the text that carefully!)

5) Could they have a few examples of ANOVA write-ups on moodle?
Would you settle for one, for now?

ROOM FOR REVIEW SESSION

We have a room reserved for today, Friday from 12-1. Please meet in Room 120 in the SOP building. Please bring questions you have about the material that will be on Exam 1!

Friday Review Session

Sorry for the late notice. We'll meet at 12 noon on Friday for the review session. We don't have a room secured yet, so we'll meet in the breezeway first. If no one is there, then stop by the front desk and we'll leave a note with the room number.

UPDATE: We will no longer be meeting in the breezeway first.

PG851 Review Session

There will be two review sessions for the first exam in PG851. One will be on Friday and the other on Tuesday. Please go to this site and indicate all the times at which you COULD come to a review session.