The Ultimate Guide to Qualitative Research - Part 2: Handling Qualitative Data

% complete
Get hands-on guidance for working with qualitative data: discover how to transcribe, organize, code, and analyze.
Team ATLAS.ti
The Experts Behind the #1 Qualitative Data Analysis Software
  1. Handling qualitative data
  2. Transcripts
  3. Field notes
  4. Memos
  5. Survey data and responses
    1. Introduction
    2. Understanding qualitative research in surveys
    3. The nature of data in surveys
    4. Transferring survey data into records
    5. Understanding survey responses
    6. Managing and storing survey data
  6. Visual and audio data
  7. Data organization
  8. Data coding
  9. Coding frame
  10. Auto and smart coding
  11. Organizing codes
  12. Qualitative data analysis
  13. Content analysis
  14. Thematic analysis
  15. Thematic analysis vs. content analysis
  16. Narrative research
  17. Phenomenological research
  18. Discourse analysis
  19. Grounded theory
  20. Deductive reasoning
  21. Inductive reasoning
  22. Inductive vs. deductive reasoning
  23. Qualitative data interpretation
  24. Qualitative analysis software

Survey data and responses

To analyze survey data, it is first important to take into consideration the process of organizing your data into a form that facilitates analysis. The analysis method most appropriate for your research will depend on the research inquiry you are looking to address.

You also need to look at how responses are structured before you can start coding or statistical analysis. Keeping all of these in mind will ensure the success of your survey research project.

There are a process and a set of considerations when you want to analyze survey data.

Understanding qualitative research in surveys

When we discuss or collect data for survey research, it's essential to distinguish between two main methodologies: qualitative and quantitative research. Both approaches offer unique strengths and can often complement each other in a mixed methods study.

However, each approach requires distinct strategies for data collection, analysis, and interpretation.

Qualitative vs. quantitative approach in surveys

In a nutshell, quantitative research involves numerical data and statistical analysis. It is typically used when researchers want to count frequency, categorize data, or measure things in a way that allows for generalizable, statistical analysis.

Quantitative survey analysis often finds insights in the statistical significance of numerical values, where differences in values between two items (e.g., the popularity of one genre of music over another) are significant enough to be confident in assertions about the survey population.

Research findings can arise from quantitative or qualitative survey analysis. Photo by Towfiqu barbhuiya.

Qualitative research, on the other hand, is non-numerical and often exploratory. It aims to delve deeper into complex issues, exploring meaning, experiences, or descriptions. Qualitative survey questions often come in the form of open-ended questions, which allow survey respondents to provide unique and individual responses. This kind of data can provide a richness of context, emotion, and depth that is not typically found in numerical data.

Ideally, survey analysis that adopts both quantitative and qualitative methods can prove useful in getting a more holistic view of the target audience and the research inquiry you are looking to address.

The value of qualitative data in surveys

The value of qualitative responses in surveys is in their depth, detail, and ability to provide a nuanced understanding of complex issues. It offers insights into participants' attitudes, behaviors, and experiences in their own words. These insights can be particularly useful in identifying patterns or themes that might not be evident from purely quantitative data.

For example, a quantitative survey might identify that a large number of employees in a company are not satisfied with their work. Still, it is the qualitative data that provides the reasons behind this dissatisfaction - perhaps there are issues with management, workload, or lack of career progression opportunities.

This is not to say that qualitative research is "better" than quantitative - they each have their own strengths and can be incredibly powerful when used together. The important thing, when it comes time to analyze survey data, is to choose the right approach for the research questions you are seeking to answer.

Definition of key terms

In order to best understand and engage with the content of this chapter and prepare for survey analysis, it's necessary to define a few key terms.

These definitions will provide a conceptual framework for our discussions on survey data collection and analysis.

What is survey data?

Survey data refers to the information or responses collected from individuals through a survey. This data can be both qualitative and quantitative.

Data from qualitative responses typically include open-ended responses, descriptions, and narratives. In contrast, quantitative data consists of numerical responses or information that can be categorized or ranked before analyzing survey data.

Survey data is a valuable resource for researchers, businesses, and policymakers, offering insights into the behaviors, attitudes, preferences, or characteristics of a sample group or population.

What is survey response analysis?

Analysis of survey responses is the process of examining, interpreting, and reporting the data collected from a survey.

This process involves a variety of techniques and approaches depending on the type of data in order to draw meaningful conclusions about how respondents answer.

Survey response analysis of customer feedback, for example, looks for customer insights directly embedded in survey results as well as how answers are framed in order to identify useful data points about market trends and consumer preferences.

Market researchers analyze survey results to identify the preferences and experiences of their target audience. Photo by Clay Banks.

For qualitative data, analysis often involves processes such as coding, thematic analysis, and narrative interpretation to understand the themes and patterns within the responses.

For quantitative data, statistical analysis methods are often used to summarize, describe, and compare the data.

What is survey data analysis?

Survey data analysis is a term that is often used interchangeably with survey response analysis. It refers to the survey analysis methods and techniques used to process, interpret, and draw conclusions from the data collected in a survey.

The type of analysis sought depends in part on whether the inquiry is qualitative or quantitative in nature. A qualitative survey analysis looks to uncover themes and patterns among survey results, while a quantitative analysis seeks out statistically significant differences among responses from different groups of survey respondents.

The goal of survey data analysis is to transform raw data into meaningful information that can be used to make informed decisions, develop strategies, or contribute to academic knowledge. Depending on the research questions and the nature of the data, different methods of analysis can be applied.

The nature of data in surveys

One of the key elements in survey research is the type of data being collected. The data collected from a survey can greatly vary depending on the survey's purpose, target population, and research questions.

Understanding the different ways to collect survey data is fundamental in designing effective surveys and efficiently analyzing the responses.

The method of data collection, whether through telephone or online survey, can affect your survey analysis. Photo by Berkeley Communications.

What kind of data is collected in a survey?

In general, survey data can be categorized into four main types: demographic, behavioral, attitudinal, and relational.

Demographic data

Demographic data provide information about the respondent's characteristics, such as age, gender, race, income, education level, and employment status. This type of data is often used to analyze and compare responses across different demographic groups.

Behavioral data

Behavioral data involves information about the respondent's actions and behaviors. This could include their purchasing habits, use of services, or lifestyle behaviors. Behavioral data can offer valuable insights into what respondents do, helping researchers understand patterns and tendencies in certain populations.

Analysis of behavioral data provides actionable insights on people's purchasing habits and lifestyle behaviors. Photo by CJ Infantino.

Attitudinal data

Attitudinal data refers to information about a respondent's attitudes, beliefs, and opinions. This data can provide insights into how respondents think or feel about specific issues, brands, policies, or services.

Attitudinal data is often collected through Likert-scale questions or open-ended questions in a survey.

Relational data

Relational data provides information about the relationships between respondents and other entities or individuals. This might include their relationship with their employer, their engagement with brands, or their interactions with public services.

Relational data is one of the types of survey data exploring connections between people and groups. Photo by Priscilla Du Preez.

Each of these types of data contributes a piece to the puzzle, helping researchers gain a more comprehensive understanding of their target population.

Understanding the diversity of survey data

While the four categories mentioned above provide a simplified overview of the types of data collected in a survey, it's important to acknowledge the diversity within this data.

For instance, within attitudinal data, researchers could be exploring a wide range of attitudes, from political opinions to consumer preferences. Similarly, behavioral data could span from online browsing habits to physical exercise routines. Each survey is unique and will collect a specific mix of data depending on its individual objectives and research questions.

Transferring survey data into records

Once data has been collected from a survey, the next crucial step is to organize this data into a format that can be easily analyzed. This process involves turning survey data into records, an important process for managing and manipulating the data effectively.

The importance of turning survey data into records

Creating records from survey data allows researchers to systematically organize, categorize, and store responses. This process enables easier access to data and facilitates its analysis. When data is appropriately recorded and organized, researchers can more effectively track patterns, identify trends, and derive meaningful insights.

How to structure survey data for recording

The structure for recording survey data largely depends on the type of data collected. Quantitative data, being numerical, is often recorded into structured formats like spreadsheets or databases, where each respondent's answers are stored in a separate row and each question in a separate column.

For qualitative data, such as responses to open-ended questions, the recording process can be a bit more complex. Responses are typically transcribed verbatim and then organized into a format that allows for text analysis, such as spreadsheets or text documents.

These formats could include coding systems or qualitative data analysis software. It's also important to note any non-verbal cues or observations if the survey was conducted face-to-face.

Categorizing qualitative responses

A significant aspect of recording qualitative data involves categorizing responses. Researchers may begin with broad categories based on the survey questions and then develop more specific categories or themes as they familiarize themselves with the data. This process, known as coding, is a crucial step in preparing data for analysis.

Data transcription and its significance

Transcription refers to the process of converting spoken language into written text or transforming written responses into a digital format. For qualitative surveys conducted in person or over the telephone, this often means typing up responses to open-ended questions, focus group discussions, or interview responses.

Survey data collected in person or over the phone may require transcription of audio recordings to generate survey results. Photo by Kelly Sikkema.

For surveys that are conducted in a digital format, there is also the great advantage that participants’ responses are already typed out and thus do not require transcription. Data transcription is an essential part of data preparation as it ensures all information is in a format that can be easily analyzed. Depending on the size and scope of the survey, transcription can be a time-consuming process. However, the benefits of having all data in a consistent, analyzable format make it a crucial step in the survey research process.

Understanding survey responses

Once survey data has been properly recorded, the next step is to understand the responses. This process involves closely examining the responses and identifying meaningful patterns, trends, and insights.

Ultimately, a critical examination of the survey results before fully analyzing data will help inform the findings in the survey report.

What are good responses?

Good responses are those that provide valuable and insightful information in relation to the survey's research objectives. While the exact characteristics of a "good" response can vary depending on the survey's purpose, there are a few common features that typically indicate a high-quality response:

Relevance: The response directly addresses the survey question and stays on topic.
Completeness: The respondent provides a full and thorough answer to the question.
Clarity: The respondent's answer is clear and easy to understand.
Detail: The response provides enough detail to give a nuanced understanding of the respondent's perspective.

Characteristics of useful responses

Beyond the qualities mentioned above, useful responses often contain insights that illuminate the respondent's perspectives, experiences, or behaviors. These might include explanations for their attitudes or behaviors, personal experiences that illustrate their point of view, or suggestions for improvements or changes.

Handling incomplete or vague responses

Inevitably, you'll encounter incomplete or vague responses in your survey data. These responses can be challenging to interpret and analyze, but they're a common part of the data collection process.

When dealing with incomplete responses, it's important to handle these in a way that maintains the integrity of your data. If a response is incomplete, it may be best to exclude it from certain analyses where it could skew the results.

For vague responses, you might have to infer the respondent's intended meaning based on the context of their other responses or categorize these responses separately during your analysis.

Validating survey responses

One of the critical aspects of managing responses is ensuring their validity. This process, known as data validation, checks that the responses are accurate, reliable, and fit for their intended use.

What is data validation?

Data validation is a process of checking the quality and accuracy of data before it's used for analysis or decision-making. In the context of surveys, validation involves ensuring that the responses are consistent, complete, and reliable.

This process may involve checking for any discrepancies or errors in the data, ensuring responses are consistent across similar questions, and verifying that the data adheres to the required format.

Why is validating responses important?

Validating responses is crucial for several reasons. Firstly, it ensures the integrity of your data, providing confidence that your findings and conclusions are based on accurate and reliable information.

Secondly, it helps identify any errors or inconsistencies in the data early in the process, preventing potential issues during analysis. This is particularly important for larger surveys, where errors can significantly impact the results.

Validating survey results is important to ensuring the quality and accuracy of survey data. Photo by John Schnobrich.

Moreover, it facilitates the use of survey data analysis methods. Responses should, as best as possible, be properly formatted and organized into a structure allowing for easy and efficient survey analysis later.

Finally, validation can also provide insights into the quality of your survey design. If many respondents are skipping certain questions or providing inconsistent responses, it may suggest that these questions are confusing or poorly designed.

How to validate responses?

There are several strategies you can employ to validate your responses:

  • Consistency checks: Compare responses to similar or related questions to check for consistency. If a respondent provides conflicting answers, it could indicate a misunderstanding or error.
  • Range checks: If your survey includes numerical responses, check that these fall within a reasonable or expected range. Any outliers may require further investigation.
  • Completeness checks: Review your data for any missing or incomplete responses. Depending on the nature of the missing data, you may decide to exclude these responses from your analysis or use statistical methods to impute the missing values.
  • Coding checks: If you've coded your responses (particularly for open-ended questions), review a sample of these to ensure the coding is accurate and consistent.

Remember that while data validation is a crucial step, it's not foolproof. It's always important to interpret your survey results with an understanding of the potential limitations and sources of error in your data.

Managing and storing survey data

After data collection and validation, a critical step is the proper management and storage of survey data. Adequate data management ensures that data remains accessible, secure, and reliable throughout the research process.

Importance of data management in survey research

Data management involves a host of activities, including data entry, storage, backup, and security. Good data management practices are essential to maintain the integrity of your research data and ensure its availability for current and future use.

Effective data management can enhance the efficiency of your research process, reduce the risk of data loss, and protect your data from unauthorized access. Additionally, proper data management can also make data sharing and collaboration easier if needed.

How to organize your survey data for easy retrieval?

When managing survey data, organization is key. Good data organization makes it easier to navigate your data, identify specific subsets of data, and streamline the data analysis process.

Here are a few strategies for organizing your survey data:

  • File naming conventions: Use consistent and descriptive file names to help you identify what each file contains at a glance.
  • Folder structures: Use a logical folder structure to organize your data files. This could be based on the survey round, data type, or any other system that suits your project.
  • Metadata: Keep a record of metadata - information about your data. This could include details about when and how the data was collected, who collected it, what each variable represents, and any coding or transformation that has been applied to the data.

Data security and privacy considerations in survey research

Given the sensitive nature of some survey data, it's crucial to ensure your data is stored securely and that respondent privacy is maintained. This involves protecting your data from both physical and digital threats.

Digital data should be encrypted and protected by strong, unique passwords. Physical data, such as printed surveys or interview transcripts, should be stored securely, and access should be restricted to authorized individuals.

In addition, it's important to adhere to relevant data privacy laws and regulations and to anonymize your data where appropriate to protect respondent confidentiality.