Survey design Flashcards

Question

What are Survey Instructions, and why are they important?

Answer 1

Survey instructions provide detailed guidance on how to complete the survey, particularly if it includes unfamiliar formats or questions. They clarify: * How to respond to different question types (e.g., multiple-choice, Likert scales). * Navigation tips for online surveys. * Submission instructions.

Answer 2

The questionnaire is the core document that contains the actual survey questions. It should be well-structured and designed to gather relevant data. Important considerations include: * Using a mix of open-ended and closed-ended questions. * Avoiding biased or leading questions. * Ensuring clarity to avoid confusion.

Answer 3

After conducting a trial run (pilot testing), feedback helps refine the survey before it is distributed on a larger scale. This ensures: * The questions are clear and easy to understand. * The survey is of an appropriate length. * Any confusing or unnecessary questions are identified and revised.

Answer 4

* **Improving Participation:** Cover letters and clear instructions make respondents more likely to engage with the survey. * **Ensuring Ethics and Compliance:** Informed consent forms protect the rights of participants. * **Improving Data Quality:** Pilot testing ensures that the survey collects accurate and relevant data.

Answer 5

1. **Define the Objective:** Clarify what you want to learn from the survey. 2. **Identify the Target Audience:** Choose the right group to survey. 3. **Design Questions:** Create clear and relevant questions. 4. **Pilot Testing:** Test the survey on a small group to ensure everything works smoothly before launching it widely.

Answer 6

* Open-ended questions * Closed-ended questions

Answer 7

**Open-ended Questions:** * Advantages: Provides in-depth and varied responses, allowing for more nuanced feedback. * Disadvantages: More difficult to analyze due to the lack of structure and the variety of responses. **Closed-ended Questions:** * Advantages: Easier to analyze and standardize responses. * Disadvantages: Limits respondents to predefined options, which may not capture their full thoughts.

Answer 8

* Use **Open-ended questions** when you want qualitative insights or to explore a subject in more detail. They are good for understanding reasons, motivations, and personal opinions. * Use **Closed-ended questions** when you need quantifiable, structured data that can be easily categorized and analyzed. These are helpful for quickly measuring satisfaction levels, preferences, or specific behaviors.

Answer 9

* Keep it simple and clear: avoid jargon and ambiguity. * Avoid leading or biased questions: will result in skewed results as respondents were compelled to respond in a certain way. * Balanced scale questions: When designing scale-based questions (such as Likert scales), it’s important to provide evenly distributed response options that cover both positive and negative sides of the scale.

Answer 10

* Nominal Scale * Ordinal Scale * Interval Scale * Ratio Scale

Answer 11

This is the most basic scale, where you just name or label things, with no ranking or order. | Just for categorising things, theres no right or wrong order.

Answer 12

This scale puts things in a specific order, but you don't know how big the difference is between them. You can rank things from best to worst, but you don’t know exactly how much better one is compared to another.

Answer 13

This scale lets you measure things in equal steps (like a thermometer), but there’s no true zero. Zero doesn’t mean "nothing"; it’s just a point on the scale. Example: "What’s the temperature today?" Options: 10°C, 20°C, 30°C. Key Point: The difference between 10°C and 20°C is the same as between 20°C and 30°C, but 0°C doesn’t mean "no temperature."

Answer 14

This is the most powerful scale because it has equal intervals and a true zero, meaning zero really means "nothing." This allows you to do math, like multiplying or comparing. Example: "What is your monthly income?" Options: $0, $1,000, $2,000, etc. Key Point: Zero means "nothing," and you can say things like "I earn twice as much as someone else."

Answer 15

Samplin is the process of choosing a subset of individuals from a larger population to represent the whole group.

Answer 16

There are 3 main types of samping: * Random sampling * Stratified sampling * Convenience sampling

Answer 17

Random Sampling means that everyone in the population has an equal chance of being selected for the survey. It’s like drawing names out of a hat, where each person is equally likely to be chosen. * **Example:** If you have 1,000 customers, you randomly select 100 of them to take your survey. * **Why it’s good:** It reduces bias and gives a fair representation of the whole population.

Answer 18

In Stratified Sampling, you divide the population into groups (called “strata”) based on certain characteristics (like age, gender, income), and then you randomly select people from each group. This ensures that each group is fairly represented in the sample. * **Example:** If your population has 60% men and 40% women, you ensure that your sample also has 60% men and 40% women, by dividing the population into male and female groups and then selecting random people from each. * **Why it’s good:** It helps ensure that important subgroups are properly represented, especially if you know certain characteristics matter for your survey results.

Answer 19

Convenience Sampling means choosing people who are easiest to reach or most readily available, rather than selecting randomly or ensuring balanced representation. * **Example:** Surveying people who walk into a specific store or sending the survey to employees in one department because they are the easiest to contact. * **Why it’s used:** It’s quick, easy, and cheap to do. * **Downside:** It can lead to bias because the people who are easiest to reach might not represent the entire population accurately.

Answer 20

* Selection bias * Response bias * Non-response bias

Answer 21

Selection Bias happens when the people chosen to take the survey are not representative of the entire population. This usually occurs because some groups are more likely to be selected than others, which can skew the results.

Answer 22

Response Bias occurs when people don’t answer questions honestly, either because they want to look good (social desirability bias) or because the way the question is worded nudges them toward a particular answer.

Answer 23

Non-response Bias occurs when certain groups of people don’t respond to the survey, leading to an overrepresentation of those who do respond. The opinions of non-respondents might be different from those who answered, skewing the results.