Week 3: Survey Design PDF

**Week 3: survey design**  Chapter 5: survey research  - Survey design involves selecting the most appropriate methods for collecting data, including structured interviews and self-administered questionnaires. **There are 2 survey formats:** - Interviews: they provide greater control...

**Week 3: survey design**  Chapter 5: survey research  - Survey design involves selecting the most appropriate methods for collecting data, including structured interviews and self-administered questionnaires. **There are 2 survey formats:** - Interviews: they provide greater control and allow for probing but can introduce interviewer bias - Questionnaire: allow anonymity and standardization but lack interaction with respondents **Types of questions:** - **Open-ended questions:** allow respondents to answer in their own words. - Useful for exploratory research or uncovering unexpected insights - Challenging to code and analyze for large-scale quantitative studies - **Close-ended questions:** provide predefined options for respondents. - Easier to standardize and analyze but restrict the depth of responses - Reduce variability and make data comparison more straightforward **Challenges in question design:** - Avoid ambiguous wording and double-barreled questions - Maintain brevity to sustain respondent focus - Place sensitive questions later to build rapport - Consider response patterns and social desirability biases **Survey administration:** - **Telephone interviews:** quick and cost-effective but lack depth - **Face-to-face interviews:** richer data but more resource-intensive - **Online surveys:** convenient but face engagement issues **Error reduction:** - Pre-test questions to minimize recall bias and ambiguity - Use vignettes for honest responses on sensitive topics - Consider secondary analysis, acknowledging potential limitations in existing data Chapter 7: quantitative sampling - Chapter 7 shifts focus to quantitative sampling techniques, crucial for selecting representative samples that allow researchers to generalize findings to a larger population.  - Sampling involves selecting a subset of the population for study, as it is usually impractical to study the entire group.  - Using proper sampling techniques ensures that inferences made from the sample can be generalized to the broader population, minimizing bias and enhancing the validity of the findings. - One of the most important techniques in quantitative research is **probability sampling.** **Probability sampling: ** - Involves random selection to give every member of the population a known and equal chance of being included in the sample.  **Types of probability sampling include**: - **Simple random sampling:** where each individual has an equal chance of being chosen - **Systematic sampling:** where every kth individual from a list is selected.  - **Stratified sampling:** where the population is divided into subgroups based on characteristics like age or gender, and random samples are taken from each subgroup.  - Multi-stage cluster sampling takes a broader approach, randomly selecting groups first, such as schools or neighborhoods, and then randomly selecting individuals within those groups. - In contrast, **non-probability** sampling methods are less rigorous and do not involve random selection.  - These techniques are often used when probability sampling is impractical.  - **Convenience sampling** relies on participants who are easiest to access, but this method can introduce significant biases, such as non-representative samples or self-selection bias.  - **Snowball sampling** is useful for hard-to-reach populations, where initial participants refer others to join the study.  - While this can help access otherwise difficult-to-reach groups, it also creates dependencies between participants and may introduce bias.  - Lastly, **quota sampling** involves selecting participants non-randomly based on specific quotas that reflect certain population characteristics, like gender or age, but it can still lead to non-randomness in other characteristics. - **Sampling challenges** are also discussed, such as the potential for bias when random sampling is not used or when sampling frames (the list of individuals from which samples are drawn) are incomplete or outdated. - Furthermore, **sampling error** can occur if the sample doesn't accurately reflect the characteristics of the population.  - The chapter highlights the importance of selecting an **appropriate sample size. ** - **Larger sample size**s generally improve the reliability and accuracy of the findings but also come with increased costs and resource demands.  - Additionally, higher response rates are emphasized as crucial to ensuring that the sample accurately represents the population, and strategies such as follow-up surveys or incentives may be necessary to improve response rates. **Lecture on this week:** - Lecture 3 introduces fundamental concepts regarding populations and sampling.  - A **population** is defined as the entire group that a researcher wishes to study, such as all students at a particular university or all residents of a country.  - A **sampling element** is a single case within the population, such as an individual participant in a survey.  - Given that it's often impractical to study every member of a population, researchers select a **sample**, which is a subset that is representative of the larger group.  - This concept connects directly to Chapter 7's discussion of probability sampling, where the goal is to select a sample that can accurately represent the broader population. - The lecture emphasizes the importance of defining the population clearly.  - A study on married couples, for example, should not include unmarried individuals, and a study focusing on teenagers should avoid including other age groups.  - The selection of a representative sample ensures that findings can be generalized to the population.  - Probability sampling methods, such as **simple random sampling**, ensure that each member of the population has an equal chance of being selected, reducing bias and improving the generalizability of the study\'s results.  - The lecture also touches on the practical limitations of sampling, where, in many cases, non-probability sampling methods, such as **convenience sampling**, are used due to time or resource constraints, though these come with the risk of bias. - In discussing **non-probability sampling**, the lecture highlights the challenges and trade-offs involved.  - While these methods may be more practical, they are less reliable in terms of representativeness.  - This connects with Chapter 7's exploration of different non-probability methods, such as **snowball sampling** and **quota sampling**, where participants are selected based on availability or specific demographic criteria.  - The lecture reinforces the importance of careful consideration of bias when using these methods.  **Connection between lecture and chapters:** - Both Chapter 5 and Chapter 7, as well as the lecture, emphasize the importance of careful survey and sampling design in ensuring reliable and generalizable findings.  - Chapter 5 discusses various survey formats, including the pros and cons of interviews versus questionnaires - While Chapter 7 focuses on the methodologies of sampling that ensure a representative sample.  - The lecture connects these ideas by explaining how to define the population clearly and select a representative sample, underscoring the need for randomness in probability sampling methods to minimize bias.  - Each section highlights challenges in survey design, such as response bias, question clarity, and sampling errors, and stresses the importance of careful planning, pre-testing, and minimizing biases to ensure the accuracy and relevance of research results.