prev

next

of 28

View

4.114Category

## Design

Embed Size (px)

DESCRIPTION

- 1. Sampling in Qualitative andQuantitative ResearchA practical how-to

2. Key themes A famous sampling mistake Quantitative assumptions in sampling Qualitative assumptions in sampling Types of sampling Ethnographic sampling Interview sampling Content analysis sampling How many? 3. A famous sampling mistake 4. A famous sampling mistakeThats Truman They only asked rich, whitepeople with telephones whod they vote for. Sadly,they published their mistake 5. Even with proper samplingbeware!predicting behavior on the basis ofknowledge of attitude is a very hazardous venture. Meaning, predicting socialbehavior is often misguided. Keep that inmind! 6. What exactly IS a sample? 7. What exactly IS a sample? A subset of the population, selected byeither probability or non-probability methods. If you have a probabilitysample you simply know the likelihood of any member of the population beingincluded (not necessarily that it is random. 8. What do quant researchers worryabout?I really spend a lotI want to know what of time wonderingI want to make surecauses somethinghow to measure I wonder how small others can repeatelse. things. patterns generalize to big my findings. patterns. 9. Assumptions of quantitative samplingWe want to generalize to thepopulation.Random events are predictable.We can compare random events Thereforeto our results.Probability sampling is the bestapproach. 10. What do qual researchers worry about?I want to see the I want to describe I really want myworld through the the context in a lot I want to show howresearch approacheyes of myof detail. social change occurs. Im to be flexible andrespondents. interested in how thingsable to change. come to be. 11. Assumptions of qualitativesamplingSocial actors are not predictablelike objects. Randomized events are irrelevant to social life. Probability sampling is expensiveTherefore and inefficient.Non-probability sampling is thebest approach. 12. Types of samples 13. Simple Random Sample1. Get a list or sampling framea. This is the hard part! It must not systematically exclude anyone.b. Remember the famous sampling mistake?2. Generate random numbers3. Select one person per random number 14. Systematic Random Sample1. Select a random number, which will be known as k2. Get a list of people, or observe a flow of people (e.g., pedestrians on a corner)3. Select every kthpersona. Careful that there is no systematic rhythm to the flow or list of people.b. If every 4th person on the list is, say, rich or senior or some other consistent pattern, avoid this method 15. Stratified Random Sample1. Separate your population into groups or strata2. Do either a simple random sample or systematic random sample from therea. Note you must know easily what the strata are before attempting thisb. If your sampling frame is sorted by, say, school district, then youre able to use this method 16. Multi-stage Cluster Sample1. Get a list of clusters, e.g., branches of a company2. Randomly sample clusters from that list3. Have a list of, say, 10 branches4. Randomly sample people within those branchesa. This method is complex and expensive! 17. The Convenience Sample1. Find some people that are easy to find 18. The Snowball Sample1. Find a few people that are relevant to your topic.2. Ask them to refer you to more of them. 19. The Quota Sample1. Determine what the population looks like in terms of specific qualities.2. Create quotas based on those qualities.3. Select people for each quota. 20. The Theoretical Sample 21. What about generalizing?Our findings have a margin of The average man is 35% moreerror of + or - 4%, 19 times out likely to choose this option of 20. over the average woman. 22. Proviso in non-probabilitysampling: no generalizingOur findings have a margin of The average man is 35% moreerror of + or - 4%, 19 times out likely to choose this option of 20. over the average woman. 23. Ethnographers samplePeople PlacesContexts TimesEvents 24. Interviewers samplePeople Places Times 25. Content analysts sampleMediaDates 26. How many? Qualitative researchers seek saturation How many isnt the issue. Do you understand the phenomenon? Have you learned enough? Mere numbers are irrelevant. You want verstehn or deep understanding Quantitative researchers seek statistical validity Can you safely generalize to the population? Have you systematically excluded anyone? (See the famous sampling mistake!) 27. Improving Response Rates Personalize the invitation Offer money -- no strings attached! 28. Copernicus ConsultingDesign research and strategy http://www.copernicusconsulting.net