S6 - Quantitative Research: Planning and collecting Flashcards
Why is it harder for descriptive research to create a sample structure?
Because the sample has to be representative of the target population in order to reduce the sampling error.
What is the difference between proportionate stratified and disproportionate stratified?
Proportionate stratified: Picking some from each subgroup however their number in the sample is proportionate to their number within the population for instance if we have more males than females in our population we will sample more males than females
Disproportionate stratified: If one group shows more heterogeneity (diversity) we would select more from it compared to a group with less heterogeneity. Allocates samples unequally, often to overrepresent smaller or specific strata for detailed analysis.
What is the difference between probability and non-probability samples? Describe the different methods.
- Probability: Every member of the target population has a known, non-zero chance of being selected. This method ensures randomness and reduces selection bias, making the sample representative of the population.
- Simple Random: Each individual in the population has an equal chance of being selected. This method is often done using random number generators or lottery systems.
- Stratified: The population is divided into subgroups (strata) based on shared characteristics, and random samples are drawn from each stratum. This ensures representation from all key subgroups.
- Cluster: The population is divided into clusters (e.g., geographic areas or schools), and entire clusters are randomly selected. This is often used for practical and cost-efficient sampling.(Ex.: People from montreal vs outside of montreal, pick only people from montreal) - Non-probability: Not every member of the population has a chance of being included. This method is often quicker and more convenient but may lead to bias and less generalizable results.
- Convenience: Samples are selected based on ease of access or availability. For example, surveying people who are readily available at a shopping mall.
- Judgement: The researcher selects participants based on their judgment of who would provide the best information. This relies on expertise but can introduce bias.
- Snowballing: Existing participants recruit others from their network, often used when studying hard-to-reach populations, such as individuals in niche communities.
- Quota: The researcher sets specific quotas (e.g., 40% male, 60% female) and selects participants non-randomly until the quota is filled. It ensures certain characteristics are represented but lacks randomness.
What are the three sampling size determination methods?
- Comparability
- Budget availability
- Statistical method
What are the two types of errors?
- Sampling errors
- Discrepancy between statistical and true value of population - Non-Sampling errors
- Due to errors in conception, logic, interpretation of responses, statistics, tabulations, reporting
Describe the procedure for developing a questionnaire.
- Develop question topics (ex.: Evaluate the preference for each flavor, Estimate purchase of intent for each flavor, Determine which one has the most potential)
- Select question and response format
- Open ended questions (ex.: What are all the brands that you are aware of?)
- Closed ended questions (Dichotomous (Yes/No) or Multi-response
- Nominal, ordinal or scale - Select wording
- Determine Sequence
- Design layout and appearance
- Pilot test
- Undertake the survey
How can you use the qualitative research findings to create your survey?
If a programme of exploratory primary research has been undertaken prior to the quantitative research, the findings can be used to clarify what should be asked and also the best ways of tackling each topic.
The qualitative research should also help in determining the most relevant wording for questions and the statements to be used in rating scales.
