Data 101: Collection & Analysis
Now that you know a little bit more about what data is from our first Data 101 post, it's time to get into the specifics: collecting and analyzing your data.
What type of data you are going to collect is perhaps the most daunting question—once you have this figured out, things get a lot easier. The type of data you need depends on:
First and foremost, what is your research question? For example, it may be that you are trying to analyze if sugar impacts hyperactivity in kids. When thinking about this, you will want to consider whether there is an independent variable (sugar) and a dependent variable (hyperactivity), and as a result, what data you need (e.g., children’s sugar consumption and how frequently children jump around after consumption).
Second, there are often many types of data that will help answer your research question, but which one is most feasible to collect? For instance, it may be difficult for you to measure children’s jumping patterns after sugar consumption, so instead, you could ask them to rank how hyper they felt after sugar on a scale from 1 (not very hyper) to 7 (extremely hyper). This is known as a Likert Scale.
Third, what type of data is your research question suitable for? Specifically, quantitative (e.g., numerical) or quantitative (e.g., descriptive) data? Quantitative data calls for measures and quantifiable instruments such as scales, whereas qualitative data calls for observations, interviews, or open-ended surveys. While the example of sugar’s effect on hyperactivity would call for quantitative data, a broader question such as why do parents give their children sugar may call for qualitative data.
Answering these three key questions will help you decide what type of data you need and how you are going to collect it.
Before you dive into analyzing your data it is important to determine what type of analyses you want to conduct. Qualitative data generally does not require statistical tools. Instead, researchers can search for themes in observations, interventions, or open-ended surveys and try to find patterns or repetition in the data, using existing research to guide the theme creation (i.e. a grounded theory approach). Quantitative data requires complex statistical formulas. Luckily, there are many user-friendly statistical tools that will perform these analyses for you! These include Microsoft Excel, SPSS, STATA, and more. Some common statistical analyses include correlations, which shows you the general relationship between two variables, and regressions, which shows you to what extent one variable predicts change in another variable.
Keep in mind that data collection and analysis is one of the most important elements of a research project. Take your time and be thorough. Happy collecting!