How do sociologists investigate society? » Types of question - multiple choice and scaled

What you'll learn this session

Study time: 30 minutes

The purpose and structure of multiple-choice questions in sociological research
How to design effective multiple-choice questions
The benefits and limitations of multiple-choice questions
What scaled questions are and how they work
Different types of scales used in sociology (Likert, semantic differential, etc.)
How to analyse data from multiple-choice and scaled questions

Multiple-Choice Questions in Sociological Research

Multiple-choice questions (MCQs) are one of the most common tools sociologists use when investigating society. They provide a structured way to collect data from large groups of people quickly and efficiently.

Key Definitions:

Multiple-choice question: A question format that asks respondents to select one or more answers from a predefined list of options.
Closed question: A question that limits respondents to choosing from provided options rather than answering in their own words.
Quantitative data: Numerical data that can be measured and statistically analysed.

📝 Structure of MCQs

A typical multiple-choice question has:

A clear question stem (the actual question)
A set of response options (usually 3-5)
Sometimes a "none of the above" or "other" option

Example: "Which social media platform do you use most often?"

Instagram
TikTok
Snapchat
Facebook
Other (please specify)

💡 Designing Effective MCQs

Good multiple-choice questions should:

Be clear and unambiguous
Have mutually exclusive options (no overlap)
Be exhaustive (cover all possible answers)
Avoid leading or biased language
Be relevant to research objectives

Poor design can lead to unreliable data!

Benefits and Limitations of Multiple-Choice Questions

✔ Benefits

Easy to complete: Respondents simply tick boxes
Quick to analyse: Data can be easily coded and processed
Standardised responses: Everyone answers the same questions
Large samples: Can collect data from many people
Comparability: Easy to compare results across groups
Reliability: Can be tested and repeated

❌ Limitations

Lack of depth: No detailed explanations
Limited options: May not capture all possible answers
Social desirability bias: People may choose "correct" rather than honest answers
No context: Doesn't explain why people chose certain answers
Researcher bias: Questions reflect researcher's assumptions

Case Study Focus: The UK Census

The UK Census uses multiple-choice questions to collect data about the entire population. For example, in the 2021 Census, questions about ethnicity, religion and employment status were all multiple-choice. This allows the Office for National Statistics to quickly process data from millions of households and produce reliable statistics about the UK population.

However, some groups argued that the fixed categories for ethnicity and gender identity didn't allow people to accurately describe themselves, showing a key limitation of multiple-choice formats.

Scaled Questions in Sociological Research

Scaled questions are a special type of closed question that measures attitudes, opinions, or feelings along a continuum. They're especially useful for measuring the intensity of people's views rather than just which category they fall into.

Key Definitions:

Scaled question: A question format that asks respondents to place their answer on a scale or continuum.
Likert scale: A five or seven-point scale that ranges from one extreme to another (e.g., strongly agree to strongly disagree).
Semantic differential scale: A scale with opposite adjectives at each end (e.g., boring exciting).
Interval data: Data where the difference between points on the scale is meaningful and consistent.

📊 Likert Scales

The most common type of scale, typically with 5 points:

Strongly agree
Agree
Neither agree nor disagree
Disagree
Strongly disagree

Example: "Social media has a positive impact on young people's mental health."

📈 Semantic Differential

Uses opposite word pairs with points in between:

How would you rate your school experience?

Boring 1 2 3 4 5 Exciting

Stressful 1 2 3 4 5 Relaxed

Unfriendly 1 2 3 4 5 Friendly

This helps measure feelings that might be hard to express directly.

📉 Numerical Scales

Ask respondents to rate something on a numerical scale:

"On a scale of 0-10, how satisfied are you with your neighbourhood as a place to live?"

0 = Not at all satisfied

10 = Completely satisfied

These scales are easy to understand and analyse statistically.

Analysing Data from Multiple-Choice and Scaled Questions

One of the main advantages of using multiple-choice and scaled questions is that they produce data that's relatively easy to analyse. Here's how sociologists typically work with this kind of data:

📊 Multiple-Choice Analysis

Multiple-choice data is usually analysed by:

Counting frequencies (how many people chose each option)
Calculating percentages
Creating visual representations like bar charts or pie charts
Cross-tabulating results (comparing answers across different groups)

For example, a sociologist might compare how men and women answered a question about household chores to investigate gender roles.

📈 Scaled Question Analysis

Scaled questions allow for more sophisticated analysis:

Calculating means (averages) and standard deviations
Creating composite scores from multiple related questions
Conducting correlation analysis to see relationships between variables
Using statistical tests to determine significance of findings

For example, a researcher might create an "environmental concern index" by combining responses to several Likert scale items about climate change.

Real-World Application: British Social Attitudes Survey

The British Social Attitudes Survey, conducted annually since 1983, uses both multiple-choice and scaled questions to track changes in public opinion over time. The survey covers topics like politics, religion, gender roles and discrimination.

For example, respondents are asked to rate their agreement with statements like "A man's job is to earn money; a woman's job is to look after the home and family" on a five-point scale. By tracking responses to this question over decades, sociologists have documented significant shifts in attitudes toward gender roles in British society.

This long-running survey demonstrates how standardised question formats allow for reliable comparisons across time something that would be much harder with open-ended questions.

Practical Tips for Using These Question Types

Pilot test your questions with a small group first to identify any problems
Consider using a mix of question types to get both breadth and depth of information
Be aware of question order effects - earlier questions can influence later responses
Include "don't know" or "prefer not to say" options for sensitive topics
Keep language simple and clear - avoid jargon or complex terminology
Consider your sample - ensure questions are appropriate for all respondents
Remember ethical considerations - respect privacy and avoid causing distress

Summary: Choosing Between Question Types

When designing sociological research, the choice between multiple-choice and scaled questions depends on your research goals:

📝 Use Multiple-Choice When:

You need factual information (e.g., "Which of these services have you used?")
There are clear, distinct categories
You want to classify respondents into groups
You need to collect data quickly from large samples

📈 Use Scaled Questions When:

You're measuring attitudes or opinions
You want to capture intensity or degree of feeling
You need more nuanced data than simple categories
You plan to conduct statistical analysis

Remember that both question types have strengths and limitations. Many sociologists use a combination of approaches to get the most complete picture of social phenomena. The best research design matches question types to research objectives while being mindful of practical constraints like time and resources.

🧠 Test Your Knowledge!