If you decided to do primary research, you must engage in analyzing data collected in order to evaluate and interpret it. In other words, you make sense of the data so that you can use it in your research paper. From the data collected in your primary research, you want to look for information that is supportive of any claims you make, creates a discussion point or provides information to incorporate in your paper. There are two types of data, qualitative and quantitative. How you analyze information collected in your research depends on the type of data that it generates.
Regardless of what type of primary research you do, you are left with a data set that falls under one of these two data categories. Quantitative data consists of numerical data, and it is represented in mathematical terms. Analyzing quantitative data involves statistical analysis. On the other hand, quantitative data is narrative data, and it is represented with data that is broken down into categories, patterns or themes.
Regardless of the type of data you intend to analyze, you must first evaluate it. Examining the raw data for information that is useful means you want to throw out any incomplete or faulty data. Data you want to exclude can consist of incomplete responses, data entry errors or questionable entries. You want all your data to stay at a high level of quality, so eliminating raw data that is obviously incomplete or faulty maintains the integrity and accuracy of your primary research. Once you evaluate and remove unusable data, you can move on to analyzing data.
To analyze quantitative data, some computing is necessary. While there are many types of statistical analysis, the three most common calculations are mean, standard deviation and frequency distribution. The mean is essentially an average of a set of numerical values. The standard deviation measures the distribution of responses on either side of the mean. It provides you with information over the consistency of responses to help better understand data. The frequency distribution clues you in to the frequency of any given response to give you an idea on the level of consensus in responses.
To analyze quantitative data, organization is necessary. You should aim to group data results into categories, themes or topics. Focus on one piece of data at a time to categorize it. During this type of analysis, the purpose of your primary research and what you measure and how you measure it should be considered to help make sense of and organize the data.
If interviews are part of your primary research, analyzing data from them is fairly simple. Any numerical data should be analyzed using statistical analysis that gives an overall picture of the results. Narrative data should be organized into categories, patterns or themes. Charting the narrative responses is also helpful. It gives you an idea on how to incorporate the information into your research paper. For any recorded interviews, transcribing the interview makes it easier to make sense of your data and look at it as whole.
If observations are part of your primary research, analyzing data from them is a bit trickier. During the observation process, you should have taken a good amount of notes. Before you can analyze the data you collected, you must first organize your notes by a defined set of criteria. From your organized notes, you can start to make some generalizations about the research as it relates to your topic.
If surveys are part of your primary research, analyzing data from them most likely involves both quantitative and qualitative analysis. Because of this, the first thing you want to do is to separate the raw data into two categories: numerical answers and open-ended answers. Once it is separated, you can compute any statistical analysis for the quantitative data, and categorize the qualitative data. Using a spreadsheet program to organize and analyze data from surveys that either provides numerical responses or yes/no answers is a good way to organize your raw data. The important thing is to organize your raw data into a form that you can make sense of with analysis to address the topic of your paper.
Whichever type of primary research you do, make sure you are not over-generalizing your results. Most likely, the number of people involved in your data collection is small, especially for a research paper. This makes your data set small. While this provides insight into whatever you are studying, it does not always translate as being true for a larger group of people or society as a whole. Because of this, use care in making broad generalizations that are not supported through small sample sizes in your primary research.
Primary research is a great tool for collecting data, and when it is analyzed and interpreted correctly and done ethically, it provides a valuable resource for your research.