Guidelines for Writing a Research Memo: Variables and Dataset Operationalization

Motivating the Memo

Task:

2)Write 1-4 sentences to motivate the memo. This can be used to discuss why the issue is important and, if applicable, whom it may affect.

3)Identify your target population if it is not immediately clear from the introduction.

4)Write a hypothesis statement reflecting what it is you are researching for this memo.  The statement does not need to be formally written, but your position should be clear and succinct.  You do not need to state the null.

5)Give a spoiler statement of your findings.

6)Identify the dataset, or subset of data, and variables you are using and operationalize (define & identify) the variables in the dataset that will represent the variables in your theory.

a)If the target population if it is not immediately clear from using the entire dataset, then identify that you are using a segment of the larger available data, e.g. GSS should be stated as nationally representative but a target population of “poor people” using the GSS will need to be defined and properly subset.

b)Your dataset must be one of the approved datasets listed as applicable to the project client list (see the Projects Page in ELMS).

c)You will need five variables—four independent (IV) and one dependent (DV). 

i)The DV must be a continuous (quantitative) variable, so it must be interval or ratio.  You may use an ordinal variable, if and only if, it has at least 5 response categories.

ii)The four (4) IVs may be any level of measurement.  Ordinal variables with at least 5 response categories may be used similarly to interval-ratio variables in the model.  Ordinal variables with fewer than 5 categories and all nominal variables must be dummy coded.  Individual dummy variables do not count toward your total of 4 IVs.

iii)You are may extend your Project 3 memo into Project 2 by reusing the same IV and DV and adding 3 more IVs into your analysis.  This is optional, but may allow you to include any Project 3 feedback and speed up your work progress.

iv)Recoding is useful if it matters to your target question and/or population.

When might you recode:

(1)If you have data entry errors to correct.

(2)If it makes sense for your RQ, theory, and hypotheses.

(3)To make continuous variable into an ordinal variable.  Think this one through first, as regression works easiest to interpret with interval-ratio and indicator variables.  Using an ordinal variable as your IV, DV, or both changes how you interpret findings compared to how you interpret continuous variables.  

7)Discuss each variable.

a)Summarize the concept of the variable that you will use.  Do not write the coded name from the dataset.  For example, if your variable measures religion, you might call it religion and one GSS variable would be RELIG16, but you would not type RELIG16 in your memo.

b)Summarize the question or statement from the codebook or other dataset supporting documents that describes the variable.

c)Summarize the response categories, such as a Likert scale, list the response options.

d)Describe applicable aspects of the variable: 

i)Interval & ratio variables: n, min, max, median, mean and standard deviation

Ordinal variables: n, frequencies & percentages

Your n for any variable should only count the valid n (those not missing).

Consider your missing values.  Report how many are missing—you may choose how to report it, e.g. frequency or proportion.  

ii)Describe any recoding that was performed on the variable and how it appeared after the recode.  Your reported summary statistics should be on the recoded variable, not the original.

iii)If you have a substantial amount of missing values or extreme outliers, mention if/how these values might affect your analysis and potential generalization to the population.