Objectives and Goals

What you ought to do while in graduate school

Several students asked me to put together a list of learning objectives and skills you might consider essential for professional development. I cobbled together a list below that ought to serve as a good guideline for your priorities while in graduate school. These guidelines are exactly that - guidelines - but they represent the skills that often separate the best scientists from the rest of the pack. Please consult this list often and email me suggestions if you find some details missing or incomplete. I grouped these objectives into three categories - presenting your ideas, analyzing data, and asking good questions. Each of these categories requires a little elaboration so I offer those details along with a checklist of items I think best represent them.


Science depends upon disseminating new ideas. You must present what you have done or else these new ideas lay fallow. I offer three areas to improve upon your presentation skills.

1. Writing

Writing remains the most essential skill for all professional activities. Those who can write - communicate effectively - often rise to the top. I expect students to write. There is no substitute for writing. If you want to be a better writer, write. Here are some useful resources but they all pale in comparison to just sitting down and writing. Have someone read what you write. Talk to others about your writing. Use writing to think and think while you write. Consider these resources:

2. Reading

Everyone needs to learn how to read - not just peruse words and grasp the point but read carefully and purposefully. Most students being graduate school with scarcely any reading skills. I suggest you learn how to be an exceptionally good reader. Read with a purpose and read for retention. Here are a few good resources that you might find useful:

3. Presenting

The most important skill that separates you from the masses is presenting well. Focus on learning these skills, practice them often, and strive to get better with every presentation. Purchase "The Exceptional Presenter," read it, and put those skills into action.

More Resources


The second most important skills to master is data analysis. Many people argue that data science offers young people the most opportunity for advancement. Learn to analyze data using the following tools and you might find yourself with more opportunities - financial, employment, and independence - compared to those without these skills. I divide these data analytic skills into roughly categories. The first is to learn a language. A language requires both syntax and semantics. R is a language; SPSS is a syntax. The reason why a language is superior to a syntax is that a language offers greater extensibility and wider application. R - a statistical language - allows the user to learn one syntax and run an infinite number of applications. A syntax restricts the user to only those applications written for the program. I recommend you learn a language and then learn these analytic tools:

1. Data Management

  • Transformations (Box-Cox, Bulging Rule, etc.)
  • Diagnostics (univariate, bivariate, and matrix plots)
  • Missing Data handling (especially multiple imputation)
  • Cleaning data (via apply, lapply, tapply, ddply, rbind, cbind, recode, or aggregate functions)

2. Data Reduction/Measurement Models

  • Principal Component Analysis (PCA)
  • Exploratory Factor Analysis (EFA)
  • Confirmatory Factor Analysis (CFA)
  • Latent Class Analysis (LCA)
  • Classical Test Theory (CTT)
  • Rasch Models
  • Item Response Theory (IRT)
  • Finite Mixture Models

3. Frequentist Models

  • General Linear Model
    • ANOVA (aov)
    • MRC (lm)
  • Generalized Linear Model
    • Mixed-effects models (lmer in lme4; nlme)
    • Generalized estimating equations (gee, geepack)

4. Latent Variable Models (SEM)

5. Resampling statistics (Bootstrapping)

  • Base graphics
  • Lattice
  • ggplot
  • graphviz


Research moves forward when researchers ask useful and testable questions. These questions often come after casual observations but more often they come after carefully reading the literature and searching for holes in the logic of prior work. We all need to be able to formulate good questions and some people master this skill immediately while others struggle throughout their careers. I want all MRES members to learn how to ask good questions. Question everything but do not question everything for the sake of questioning; instead, question everything and see if you might be able to address your question with data. We are empiricists. I love philosophy but data offers us a stronger method to question and address our questions. Before we can address questions, we must ask them. How do we learn to ask questions? Here are some useful steps:

1. Read.

Read. Read. There is no shortcut for learning about a new field than reading what has been done. Read critically. Read broadly. Read.

2. Discuss.

Talk to others about what you read. We have weekly MRES meetings to facilitate these discussions. Read and come talk to us all about what you read.

3. Search.

Google makes your life easier but not easy. Search for what has been done and think about what has yet to be done.

4. Challenge yourself.

Part of what makes a person good at asking questions is that they ask many questions and they get feedback about them. Ask questions to others about things you believe you know. The aim is not to patronize but rather to challenge yourself to think deeper about what you know.

5. Adopt multiple perspectives.

Every topic has multiple perspectives. Discover these different perspectives and adopt each one to ask questions that disconfirm them.

6. Consider your grandmother/grandfather.

Good questions ought to be simple. Ask questions that someone like your grandmother would understand and appreciate. The simpler you can make the question, the more likely you can remember, articulate, and address that question in your own work.

7. Learn how to learn.

Be an independent learner. Shift your content areas frequently so you may gain a broader understanding of your scientific discipline. Hyperspecialization may get you short-term rewards but the long-term rewards come to those who bridge multiple areas to offer new insights. Learn how to learn these areas and you set yourself up for better questions with more important implications.