Assignments
Participation in class is expected and will be scored at one point per class, for a total of 10 points, which is 10% of your total grade. There will also be three homework assignments at 15 points each (45 points), a final project proposal (5 points), and the final project itself (40 points), for 100 total points.
A note on deadlines
I would like to, as much a possible, stick to the deadlines below so we can go over them together as a group after everyone has submitted their assignment. However, if you need additional time for any reason please just send me a note letting me know. You do not need to justify why. I would just ask that you not attend class during the time we are going over the assignment (but please attend the rest of the class if you are able).
Participation
Please turn in your script from the day’s work to canvas after each class meeting. Very regularly this will include specific “challenges”, and annotating your code to note these would be helpful. Other time, I will ask you to follow along with me, and you can turn in your script from this time.
Homework
Each homework assignment is worth 15 points. Please do not turn in partial work. Instead, please ask for help and, if needed, an extension on the deadline.
Lab | Date Assigned | Date Due | Topic |
---|---|---|---|
1 | Fri, April 09 | Fri, April 23 | Basic multilevel modeling with R |
2 | Fri, April 30 | Fri, May 14 | Growth models and variance-covariance matrices |
3 | Fri, May 14 | Fri, May 28 | Bayesian estimation & multilevel logistic regression models |
Final Project
A proposal for the final project is due Friday, April 30th, before midnight. The proposal should have the following elements:
- Data source identified, which must be shareable with me
- Research Question(s) identified (no more than three), which must be addressable through a multilevel model
- A description of any data processing that must occur before you can fit the given model
- Your current status with the project (e.g., challenges you are facing, what steps still need to occur, feasibility of finishing, etc.)
The final product includes two parts: (a) an R script of your analysis and all plots, and (b) a writeup of the results of your analysis which should largely match APA style. You will also need to submit the data you are analyzing so I can run your script locally.
R Script (20 points)
The R script you submit for your final project should be a .R or .Rmd file. It should include any necessary data pre-processing steps needed to complete the analysis (e.g., moving the data from wide to long). You will not be graded on the data preparation portion of the script, but I will give you feedback.
We will discuss this more in class, but please annotate your script clearly to indicate what is happening in each section (e.g., data preparation, exploratory plots, descriptive analyses, preliminary analyses, primary analyses, model-based plots, etc.). I will score your R script based on the following criteria:
-
Reproducibility: 2 points
- I should be able to run your script without errors with little effort. The results on the analysis I run locally should match those that you report in the writeup. Note, if you are using simulation or anything relating to random processes, it will be important to set a seed to ensure reproducibility.
-
Exploratory and descriptive analyses: 3 points
- One of the most important components of analysis work generally is knowing your data well. The exploratory and descriptive analyses will inform you on whether you data meet your model assumptions and, if not, what data transformations or amendments to your model should be made. Note that it is acceptable to identify areas where the data may not meet the model assumptions, note them, and move on. In other words, you do not have to address everything you find, but they should be noted in the limitations section of your writeup.
-
Analysis: 10 points
- Depending on your specific situation, you may engage in a model building process (starting with a baseline model and adding predictors) or you may have a specific model in mind a priori that you can directly estimate. In either case, the model should map directly to the research question and be properly specified. There are often judgments that must be made in your analysis and your decisions should be clear from the evidence. You can defend these judgments in your writeup, but failure to properly evaluate important decision points (e.g., whether a variable should randomly vary or not) without theoretical justification in the writeup will result in a loss of points.
-
Plots: 5 points
- You are required to create one plot from your model that communicates the coefficients, model predictions, or other features. You are welcome to include more than one, but at least one plot should be created.
Writeup (20 points)
You will develop very short APA writeup of your model results, which should not exceed five pages, double-spaced, with standard 1 inch margins and 12pt Times New Roman font. Manuscripts that exceed the five page limit will have five points removed immediately. As indicated below, you will be required to include at least one figure and at least one table, which along with references, will not count against the page limit. The manuscript should include a (very brief) introduction, methods, results, and a brief discussion including the limitations. As the points below and suggested page limits for each section indicate, the majority of the space should be devoted to the methods and results. The writeup should include the following components:
- Introduction: 2 points
- Suggested length - one paragraph, 0.5 page max.
- Provide a very brief background on the purpose of the study.
- Research Question(s): 2 point
- Suggested length - one to three sentences.
- At least one and no more than three research questions to be addressed in the writeup, which should be addressable from the data and within a multilevel modeling framework.
- Method: 5 points
- Suggested length - 1.5 to 2 pages.
- Describe where the data came from (one paragraph)
- Describe your data preparation (e.g., handling missing data, transformations)
- Describe your analytic sample, complete with a table of descriptive statistics
- Describe the model you plan to fit and/or your model building process and why it is appropriate for your sample and research question(s).
- You may optionally include and discuss any exploratory plots here as well
- Results: 5 points
- Suggested length - 1.5 to 2 pages
- If appropriate, discuss the results of your model building process and your model comparisons.
- Include and discuss at least one table of your model results. Make sure you link specific coefficients or model results to your research question.
- Include at least one plot from your model that communicates your findings (e.g., perhaps comparing the model predictions to the raw data, or the model predictions for two or more groups)
- Discussion: 3 points
- Suggested length - 0.5 to 1 page
- Briefly note any limitations
- Very briefly mention how your findings link to the extant literature
- General style: 3 points
- The manuscript should generally be free of errors and read similarly to a manuscript that has been submitted for peer review.
- APA formatting should be used, with a formatted bibliography listing all references to the extant literature.