How to analyze survey data

You have recently started an online survey? Congratulations! The next step is to draw conclusions from a huge variety of data and you might arise the questions how to set up an effective survey data analysis. There are several ways to make sense of qualitative data or quantitative data. Thus, you can examine overreaching patterns and trends you would have missed otherwise.

What is survey data analysis?

At first, it is useful to examine what survey analysis actually is: The term survey analysis describes the process of examining results from customer or other surveys. The aim is to understand certain factors that influence a particular target group. When you are reading a newspaper article which is referring to some statistics, you might come across statements like “80 % of all customers” or similar formulation. This is a clue to survey data which had been analyzed. Be careful when you are reading articles like this because statistics always refers to a sample size and there might be a huge margin of error.

Types of survey data

There are several types of survey data and survey questions: quantitative and qualitative data for example are the most rudimentary distinction between data types. We are going to take a closer look at qualitative data types.

Close-ended questions – or just yes or no

Closed-ended questions are easy to analyze because they can be answered only by one word such as “yes” or “no”. These survey questions often come up at the beginning or the end of an interview and pretend several answers.

Open-end questions – customers are telling in their words

In contrast to close-ended questions there are open-ended questions where the interviewers are supposed to answer in more detailed ways. Questions like these are often used in Customer Behavioral Researches to outline meanings, feelings and knowledge. By analyzing data like these you will be able to improve your customer service based on the customer feedback you have received.

How ATLAS.ti can help you to analyze your survey data

For building up these statistics ATLAS.ti comes into play, which makes drawing conclusions easier. On the one hand, the software organizes all the survey responses you collected. On the other hand, it precedes the data in one click. In order to introduce you briefly to our software we show how to import data and to deal with several analyses.

Please take a closer look at your research design to ensure you are collecting meaningful data. If everything is fine, you will be able to start your survey. To do this, follow these steps

  1. Create an online survey and collect responses
  2. Download and save survey data in an Excel file
  3. Edit the column headers to tell ATLAS.ti how you want the data organized
  4. Import the responses (Excel file) into ATLAS.ti 8 Windows/Mac

The software creates a document for each participant and collects the given answers to the open-ended questions. Moreover, the software creates document groups from single and multiple choice questions to make comparisons between several positions possible. Quotations are created for each answer and coded with the respective question (you may use whichever code you wish). ATLAS.ti performances important preparatory work in a few seconds and you do not have to sort data manually.

Dive in to your analysis Many online surveys consist of a mix of closed-ended questions and open-ended questions, which are converted to a row in Excel. Each row of the Excel spreadsheet is going to be converted into an own document. If you have 30 responses, you will get 30 documents.

How to prepare survey data

By naming the column headers of your survey data in Excel spreadsheet and using specific prefixes, you can organize your data collection:


If you are analyzing a mix of open-ended and multiple-choice questions, it is useful to group the types of answers by demographic information. ATLAS.ti puts all the participants of the same demographics as age, gender, profession and so on together. To do this, enter a : (colon) in the column header of the closed-ended question in the Excel spreadsheet. For Further information on how ATLAS.ti works please read this ​article​

Survey data analysis – how it works

To determine whether the data is statistically significant, you have to check it carefully. The term statistically significant means your sample size has a certain size and your survey results are accurate within a certain confidence level. In the next paragraph we are going to point out some of these methods. Of course, the full functional range of ATLAS.ti includes many more analyzes.

There are several methods to check statistics – for example cross tabulating

Let’s imagine you are attempting to design a new correspondence course and you want to find out how important flexibility is to your potential students, professors and other staff members. To figure out, you have to group the response rate by groups. Afterwards you present the data collection in a cross tabulation:


This table shows, many survey respondents are interested in flexibility – especially students and teachers. If you want to examine this topic more carefully you will have a set of several research questions. These should help you to find out why flexibility is so important to your target group.

Narrative research is set out by the validation of the audience.

Narrative analysis and research are part of social science research, but are not always to be considered as a stand alone for evidence and support for the conclusion of a report. Whether as part of a presentation or as an independent work, narrative research has to be seen as independent research and interpretation in its own right. Although, every story has been put into its cultural context, the question arises how objective these experiences are. The results of research are influenced by the personal narratives and the varieties of research and interpretation may lead to misunderstandings. Therefore, you have to look carefully at your narrative methodologies to avoid misunderstandings.

Sometimes it is difficult to look at the story objectively. But you have also to put it into its sociocultural context. By this way your data analysis takes several varieties of narrative into consideration and a phenomenon or a story can be viewed from different perspectives. According to this highly complex process narrative methodologies are often used in social science.

Get started with Qualitative or Mixed-Method Research with ATLAS.ti 9