Quantitative “large N” research is currently the best way to make scientific inferences about large populations (people, animals, neighborhoods, etc.). Unlike qualitative “small N” research, which aims to intensely investigate the workings of a phenomenon in several cases, quantitative “large N” studies investigate phenomena among dozens, hundreds, and preferably thousands of cases, albeit less thoroughly per case than “small N” studies. Furthermore, quantitative studies rely on mathematical methods—like statistical analysis—rather than humanistic methods like interviewing (although quantitative studies might still use interviews). While “small N” studies are better able to articulate complex causal mechanisms, they are less able to make broad inferences about large populations—something one often wants to do in science.
Although many people believe statistical analysis to be the most daunting and opaque part of quantitative research, in reality the greatest bulk of total study time is spent setting up the study, gathering data, and making sure data is analysis-ready.
This course aims to teach the basics of how to structure and, to some extent, conduct a quantitative “large N” research study. It is designed for diverse audiences; graduate degrees are not necessary nor are degrees in particular fields. After completing the course, students should be able to lead simple research studies on their own and to assist in leading complex studies.
The framework presented here can be applied to any question one wants to answer with data; it need not be an academic study intended for peer-reviewed journal publication. Studies can be conducted in businesses, schools, non-profit organizations, or anywhere else where one is dealing with mathematically measurable concepts. Perhaps even more important than being able to conduct a study, this course can help students learn how to think about everyday problems more scientifically.
What am I going to get from this course?
- Understand the basics of how to structure and, to some extent, conduct a quantitative "Large N" research study
- Lead simple research studies on their own, and assist in leading complex studies
Module 1: Synopsis, Audience, and Objectives
Synopsis, Audience, and Objectives
This lecture gives an overview of what will be covered throughout the course, the background of the instructor, the target audience, and the learning objectives.
Module 2: Choosing Partners
This lecture will go over what types of different skill sets are needed in the process of conducting a large scale study.
Examples about how to choose partners in various contexts such as academic, business, and non-profit
More exercises and an answer sheet to help the audience grasp the concepts better
Module 3: Generating Hypotheses
This first lecture of the module explains the main points to consider when generating a hypothesis
This lecture provides multiple exercises to make sure you have a solid understanding of what a complete hypothesis looks like
Module 4: Sample Size
Definition and Importance
An explanation of what sample size is, and why it's important
Further Analysis and Examples
This lecture will go over some of the main methods used to determine the ideal sample size, and provides the audience with extensive exercises
Module 5: Sample Selection
Methods of Sample Selection
This lecture will go over Randomized Control Trials, Non-Randomized Control Trials, Observational Studies, and Surveys. It will also provide examples for each of these topics.
Continuing from part 1, this lecture will give a thorough explanation of the Classroom Studies method and the Experiments method. After presenting these, extensive exercises will be provided to further the understanding of the student.
Module 6: Selection Bias
Types of Selection Bias
This lecture will go over various kinds of selection bias ranging from Non-Random Sampling to Observer Selection. Example situations for each of these different types of selection biases will be provided.
How to Recognize and Eliminate Selection Bias
This lecture will give you some tips and tricks to avoid selection bias. Extensive exercises will be provided at the end of the lecture.
Module 7: Preliminary Data Work
Examining the Data Before the Analysis
This lecture is about eyeballing the data and making sure everything is in place before starting the experiment/analysis.
Examples and Exercises
This lecture will provide further explanation of different methods of data cleaning, and conclude with various types of exercises and examples
Module 8: Hypothesis Testing
Now that you have collected the data and cleaned it, it's time to test your hypothesis. This lecture will go over some of the common methods used in hypothesis testing.
This lecture will continue from where Part 1 left off, and will provide you with multiple exercises so you can assess your level of understanding
Module 9: Final Topics
Multiple Comparisons, Statistical vs. Practical Significance
This lecture will go over the problem of multiple comparisons, and why there's a difference between being statistically significant and practically significant.
Positivity Bias, Reporting Results, Graphing Rules
This lecture will go over what Positivity Bias is and how it effects our daily lives, how to report the results of a study, and the basic rules of graphing the results.