Let's talk about joins | Crystal Lewis

cghlewis.com · 4,642 words · saved by 1 readers

Working with data would be so much simpler if we always only had one dataset to work with. However, in the world of research, we often have multiple datasets, collected from different instruments, participants, or time periods, and our research questions typically require these data to be linked in some way. Yet, before combining data, it’s important to consider what type of join makes the most sense for our specific purposes, as well as how to correctly perform those joins. This blog post reviews the various ways we may consider combining our data. In general, there are two ways to link our data, horizontally or vertically. When linking or joining data horizontally we are matching rows by one or more variables (i.e., keys), making a wider dataset. When joining vertically, column names are matched and datasets are stacked on top of each other, making a longer dataset. Joins can be done in many different programs (e.g., SQL, R, Stata, SAS). Most of this post will be applicable to any la

Let’s talk about joins | Crystal Lewis Working with data would be so much simpler if we always only had one dataset to work with. However, in the world of research, we often have multiple datasets, collected from different instruments, participants, or time periods, and our research questions typically require these data to be linked in some way. Yet, before combining data, it’s important to consider what type of join makes the most sense for our specific purposes, as well as how to correctly perform those joins. This blog post reviews the various ways we may consider combining our data. Types

Let's talk about joins | Crystal Lewis

related reading