Psychology 2500 (which was labeled Psychology 3500 prior to Fall 2018) is an applied statistics class suitable for beginners with modest mathematics backgrounds as well as more mathematically inclined psychology majors.  We cover a broad range of topics including measures of central tendency and variability, basic probability, the normal and binomial distributions, hypothesis testing and statistical power, brief introductions to Bayes’ theorem and signal detection theory, means testing such as t-tests and ANOVA (one-way, repeated measures, two-way factorial), post hoc tests for deconstructing ANOVA results, Pearson and Spearman correlations, linear regression, and nonparametric methods including chi-square, Mann-Whitney, Wilcoxon, Kruskal-Wallis, and Friedman tests.

Example syllabus:  Fall 2022.

The course can be taken for either 3 or 4 credits.  The 4-credit option includes additional work and training in the use of the statistical software package R and its integrated development environment RStudio (developd by Posit) both of which are free, open-source, and cross-platform (Windows, MacOS, Linux).  There are many resources available on the internet for learning R and RStudio efficiently and effectively, including DataCamp training videos, resources and links from RStudio/Posit, support sites like R-Tutorial and R-bloggers, and numerous Q/A sites and individual postings.

PSYCH 2500 does not use or require calculus.  The gist of what statistical analysis is for will be learned on the first day; the rest of the course consists of practical details of implementation.  Students seeking a more thorough treatment of the theoretical underpinnings of statistical analysis should look elsewhere for a course that includes calculus.  Students seeking a more advanced applied course in statistics for the social/behavioral sciences (including the advanced use of R) should consider PSYCH 4750/4760, Quantitative Methods 1, 2, or other options as described below.  Students seeking an introductory course in statistics that includes calculus and linear algebra may wish to consider the Engineering College courses ENGRD 2700, Basic Engineering Probability and Statistics and ORIE 3500, Engineering Probability and Statistics II.

In addition to the Psychology major, PSYCH 2500 fulfills the statistics requirements or recommendations for other majors including Biological Sciences and Human Development.

Statistics Requirement for the Psychology Major

As noted on the departmental page for the undergraduate major, proficiency in statistics can be demonstrated in any one of the following ways:

  • Passing PSYCH 2500.
  • Passing an approved course or course sequence in statistics in some other department at Cornell (see below).
  • Passing an approved course or course sequence in statistics at some other college, university, or college-level summer school.  The description of the course from the college catalog, a PDF copy of the syllabus, and the title and author of the textbook(s) used must be submitted to Prof. Thomas Cleland for approval.  High school AP credit will not suffice.
  • Passing an exemption examination. This examination can be given at virtually any time during the academic year by mutual arrangement.  This does not mean that it will take place immediately upon demand.  Students who have completed a theoretical statistics course in a department of mathematics or engineering and who wish to demonstrate competence in applied statistics usually find this option the easiest.  Students considering this option should discuss it in advance with Prof. Cleland.

Basis for approval of other statistics courses to satisfy Psychology major requirements

Courses that satisfy the Psychology major requirement should cover most of the topics covered by PSYCH 2500, as listed below.  Courses from other departments are optimized for the analysis needs of different fields, and so may be excellent courses for those fields but not ideal for the Psychology major.  Some statistics and probability courses in the Economics department, for example are excellent but tend to focus on complex regression models, while providing less treatment of hypothesis testing or of means-testing methods such as ANOVA. These courses generally will satisfy the Psychology major requirements only for Psych-Econ double majors.  Other courses, like PUBPOL 2101 and HADM 2010, are closely tied to major-specific themes and may be limited to students in that major.

NOTE:  If a statistics course is accepted by your major adviser to meet the statistics requirement for the Psychology major, then it will automatically also count towards the 40 credits of the Psychology major and the 100 credits of Arts college coursework necessary to fulfill the Psychology major and graduate from the College of Arts and Sciences, even if the course itself is from a different college.

Approved statistics courses for Psychology

NOTE to Psychology majors:  Your major adviser is the person who officially determines whether a given course will meet the statistics requirement for the major.  Please confirm with that professor whether a course that you are considering is acceptable.  Also, course contents can vary substantially from year to year.  That said, the following are likely (not necessarily guaranteed) to meet the requirements for the Psychology major:

The following courses are not recommended for the Psychology major:

The following statistics courses are not introductory, but may be of interest once a semester of introductory statistics has been completed:

  • ILRST 2110 / STSCI 2110, Statistical Methods for the Social Sciences II.
    • This is an excellent course to take after Psych 2500, as it covers regression-based methods that we only briefly cover.
    • Teaches the use of statistical software (JMP, as of 2019).
  • ILRST 2130, Applied Regression Analysis.
    • A shorter, seven-week fall course for 2 credits.  Covers regression analysis with an emphasis on practical analyses of real datasets.
  • BTRY 3020 / STSCI 3200 / NTRES 4130, Biological Statistics II
    • Another second-semester course including regression techniques, generalized linear models (a superset of analysis of variance), multifactor ANOVA, and nonlinear modeling.  This course is good followup and complement to what we do in PSYCH 2500.  Although it is said to require calculus, I have been informed that this is not actually necessary.
  • BTRY 3080 / STSCI 3080, Probability Models and Inference
    • A calculus-inclusive second (or third) course focusing on probability and parametric inference.  Dig more deeply into core concepts and theory, and receive more of a thorough introduction to Bayesian methods than we cover in 2500 (but probably not as much as in PSYCH/HD 4750/4760).  Uses R.
  • PUBPOL 3100, Multiple Regression Analysis.
    • Multiple regression techniques from an econometrics perspective.  Potentially less applicable to psychological and related research datasets.

More hardcore options:

Data Science courses.

  • PSYCH/HD 2930, Introduction to Data Science for Social Scientists
    • An introductory data science course that uses R and emphasizes the goals and practices of the social sciences.  Includes the discovery and importation of diverse datasets, tidy data management, data visualization methods, and the use of predictive classifiers.  Probably the clearest data science followup course to PSYCH 2500.
  • PSYCH/HD 2940, Data Science for Social Scientists II
    • A followup course to PSYCH/HD 2930, covering more advanced methods and techniques.
  • STSCI 1380, Data Science for All.
    • An introductory data science course that requires only high school mathematics; no calculus or programming experience.  Might be too basic for those seeking to build on their PSYCH 2500 experience in a data science context.
    • Crosslisted as CS 1380, INFO 1380, ORIE 1380.
  • INFO 2950, Introduction to Data Science
    • This Information Science courses requires “strong performance in an introductory statistics course” as a prerequisite, along with an introductory programming course.  Our R portion may squeeze by as adequate here, but a superior option would be introductory Python (Python is used in this course).
  • There are MANY more data science-related courses at Cornell.

Some other courses of possible interest:

The following research methods courses also may be of considerable interest once a semester of introductory statistics has been completed.  These courses include statistical analysis but are intended to emphasize the other elements of research design that we discuss in PSYCH 2500 — i.e., how to design a study that avoids all the basic logical pitfalls from which no amount of statistical number crunching will save you.