STAT 313 Last Day

Deadlines

  • Final revisions are due by Sunday at midnight.
  • Final Projects (poster submissions) are due by the beginning of Saturday’s poster session (10:10am).

Deadline Extensions

You cannot request deadline extensions for the final version of your Final Project. The assignment portal closes at 10:10am on Saturday. Do not ride the line.

Peer Review Session (20-minutes)

Advice on Poster Formatting

You will likely need to cut down some of the content from your report to fit on your poster!

For people to be able to read your poster, your font size should not be any smaller than 24pt.

  • You are expected to include all the required components in your poster.
    • To fit everything, your writing needs to be clear and concise.

Getting Help from Generative AI

Humans tend to be very wordy in their writing, so this might be a good place to ask Generative AI for some help making your writing more concise.

Be sure to read the suggested edits so nothing important gets cut!

More Advice

  • I don’t need to see statements of hypotheses (e.g., \(H_0\) or \(H_A\))
    • I can infer what your null and alternative hypotheses were based on your conclusions!
  • Please don’t use the s-word (significance). 😱
    • Provide a more descriptive summary of your findings!

Structure of Final Projects

Hypothesis Test Conclusions

Conclusions should be written in terms of the alternative hypothesis


Did you reject the null hypothesis?

Then you have evidence that at least one group has a different mean!

Did you fail to reject the null hypothesis?

Then you have insufficient evidence that at least one group has a different mean!

Discussion

Describe whether you believe the tests you performed are “reliable.

  • Independence
    • within groups
    • between groups
  • Normality of the distributions for each group
  • Equal variance of the distributions for each group

Conditions for Each Test

Each one-way ANOVA test considers different groups. So, your conditions should be evaluated for each test separately.

Conditions are never met!

\(H_0\): the condition is met

\(H_A\): the condition is violated


Just like we never say “I accept the null hypothesis,” we never say a condition is “met.” Instead, we say there is no evidence that the condition is violated.

Study Limitations

This section summarizes your understanding of the foundational aspects of experimental design.

Based on the sampling method used, what larger population can you infer the results or your analysis onto?

  • What were the inclusion criteria of the observations?
  • How does that influence the population you can infer your findings onto?

Based on the design of the study, what type of statements can be made about the relationship between the explanatory and response variables?

  • Were the explanatory variables randomly assigned to control for confounding variables?
    • How does that influence what you can and cannot say about the relationships between the variables?

Overall Conclusions

Based on the results of your analysis what is your conclusion for the questions of interest? Connect your conclusion(s) to the relationships you saw in the visualizations you made and the results of your hypothesis tests.


Did you distributions look similar but your hypothesis test said at least one group was different?

Think about how sample size effects p-values!

Did you reject the null hypothesis for your one-way ANOVA?

Look back at your visualizations – which group(s) look the most different?

Did you fail to reject the null hypothesis for your one-way ANOVA?

Look back at your visualizations – do all of the groups look relatively similar?

Remedying Condition Violations

Is independence violated?

You should have only one observation per observational unit (e.g., fish, salamander, teacher).

and_vertebrates

vertebrates_small <- and_vertebrates %>%
  group_by(pitnumber) %>% 
  slice_sample(n = 1) %>%
  ungroup()

OR

vertebrates_small <- filter(and_vertebrates, 
                            year > 2007, 
                            is.na(pitnumber)
                            )

evals

evals_small <- evals %>% 
  group_by(prof_ID) %>% 
  slice_sample(n = 1) %>% 
  ungroup()

OR

evals_small <- evals %>% 
  group_by(prof_ID) %>% 
  mutate(min_score = min(score)) %>%
  distinct(prof_ID, 
           .keep_all = TRUE) %>%
  ungroup()

Do you have really skewed data?

Try using a log transformation!

Un-transformed Variances

unittype var
C 84.826493
I 80.783024
IP 5.729663
P 129.138547
R 59.096925
S 49.280157
SC 49.923399
NA 112.284569

Log Transformed Variances

unittype var
C 1.6978424
I 0.8788463
IP 0.8990321
P 1.5659500
R 1.3621461
S 2.1770549
SC 1.6514917
NA 0.7591942


What do you think? Did it work?

Final Poster Presentations

Presentation Structure

Each section will have an 80-minute poster presentation session.

  • Section 01: 10:10am to 11:30am
    • Session 1: 10:10am to 10:50am
    • Session 2: 10:50am to 11:30am
  • Section 02: 11:40am to 1:00pm
    • Session 1: 11:40am to 12:20pm
    • Session 2: 12:20am to 1:00pm

You can find what “Session” you are in on Canvas in the Final Presentation group.

Bring Your Laptop!

There are no computers in the presentation room!

Work Sesssion

Order of Priorities

  1. Fix your code (if needed)

  2. Copy content over to your poster

  3. Modify the original content based on Dr. Theobold’s feedback

Assistance from Dr. Theobold

My priority is helping students who need to repair their code for their analysis. Once everyone has been helped, I’m happy to answer additional questions.