Presentation is loading. Please wait.

Presentation is loading. Please wait.

Sampling. Why Sample? Some Issues: n Time, cost, accuracy n accuracy/ representativeness n Link to interesting general introduction of sampling for public.

Similar presentations


Presentation on theme: "Sampling. Why Sample? Some Issues: n Time, cost, accuracy n accuracy/ representativeness n Link to interesting general introduction of sampling for public."— Presentation transcript:

1 Sampling

2 Why Sample? Some Issues: n Time, cost, accuracy n accuracy/ representativeness n Link to interesting general introduction of sampling for public Link n Link to website advertising services of market research firm market research firmmarket research firm

3 What is a sample? Key Ideas & Basic Terminology n Link to good introduction to concepts & issues Link n Population, target population Population u the universe of phenomena we want to study u Can be people, things, practices n Sampling Frame (conceptual & operational issues) u how can we locate the population we wish to study? Examples: F Residents of a city? Telephone book, voters lists F Newsbroadcasts? Broadcast corporation archives? … F Telecommunications technologies?.... F Homeless teenagers? F “ethnic” media providers in BC (print, broadcast…)

4 Diagram of key ideas & terms in sampling

5 Target Population n Conceptual definition: the entire group u about which the researcher wishes to draw conclusions. n Example Suppose we take a group of homeless men aged 35-40 who live in the downtown east side and are HIV positive. The purpose of this study could be to compare the effectiveness of two AIDs prevention campaigns, one that encourages the men to seek access to care at drop-in clinics and the other that involves distribution of information and supplies by community health workers at shelters and on the street. The target population here would be all men meeting the same general conditions as those actually included in the sample drawn for the study.

6 Bad sampling frame = parameters do not accurately represent target population u e.g., a list of people in the phone directory does not reflect all the people in a town because not everyone has a phone or is listed in the directory.

7 Recall: Videoclip from Ask a Silly Question (play videoclip) n Ice Storm, electricity disruption, telephone survey n Target Population: Hydro company users n Sampling frame: unclear, probably phonebook or phone numbers of subscribers n Problem: people with no electricity not at home but in shelters n Famous examples from the past: Polls of voters before election (people with phones or car owners not representative of total voters, or opinions not yet formed)

8 More Basic Terminology n Sampling element (recall: unit of analysis) n e.g., person, group, city block, news broadcast, advertisement, type of media coverage, etc…

9 Sampling Ratio n a proportion of a population F e.g., 3 out of 7 people F e.g., 3% of the universe

10 Non-probability Sampling 1. Haphazard, accidental, convenience (ex. “Person on the street” interview) Babbie (1995: 192)

11 Non-probability Sampling 2. Quota (predetermined groups) Neuman (2000: 197)

12 Why have quotas? n Ex. populations with unequal representation of groups under study u Comparative studies of minority groups with majority or groups that are not equally represented in population u Study of different experiences of hospital staff with technological change (nurses, nurses aids, doctors, pharmacists…different sizes of staff, different numbers)

13 Non-probability Sampling 3. Purposive or Judgemental n Unique/singular/particular cases n Range of different types n Hard-to-find groups n Leaders (“success stories”) n Link to example of Ipsos Reid study on conducting business abroad Ipsos Reid study Ipsos Reid study

14 . Snowball (network, chain, referral, reputational) Non-probability Sampling 4. Snowball (network, chain, referral, reputational) n Often uses Sociograms u Link to instructions for doing sociograms instructions for doing sociogramsinstructions for doing sociograms

15 Non-probability Samples 5. Deviant case (type of purposive sampling) x x x x x x

16 New technologies mapp New technologies & mapping interactions n Data mining & the “blogosphere”) Data mining & the “blogosphere”) Data mining & the “blogosphere”) n On-line observation of social networks

17 Visualizations & sampling n Conversation Clock u Karrie G. Karahalios and Tony Bergstrom. Visualizing audio in group table conversation. IEEE TableTop2006. u Social spaces group (Illinois) Social spaces Social spaces

18 “Virtual” Communication n Visual Who project at MIT: visuals, more Visual Who visuals,more Visual Who visuals,more n Patterns of presence & association

19 IssuesIssues in Non-probability sampling IssuesIssues in Non-probability sampling n Sampling Bias n Is the sample representative? Of what? Of whom?

20 Types of Probability Sampling n 1. Simple Random Sample: link link

21 How to Do a Simple Random Sample n Develop sampling frame n Select elements using mathematically random procedure u e.g. Table of random numbers n Locate and identify selected element n Link to helpful website Link

22 2. Systematic Sample (every “n”th person) With Random StartSystematic Sample 2. Systematic Sample (every “n”th person) With Random StartSystematic Sample Babbie (1995: 211)

23 Problems with Systematic Sampling n Biases or “regularities” in some types of sampling frames (ex. Property owners’ names of heterosexual couples listed with man’s name first, etc…) n Urban studies example) rban studies example)rban studies example)

24 Other Types n Stratified Neuman (2000: 209)

25 ng Stratified Sampling: Sampling Disproportionately and Weighting Babbie (1995: 222)

26 Stratified Sampling n Used when information is needed about subgroups n Divide population into subgroups before using random sampling technique

27 Other Types (cont’d) n Cluster n When is it used? u lack good sampling frame or cost too high Singleton, et al (1993: 156)

28 Other Sampling Techniques n Probability Proportionate to Size (PPS) n Random Digit Dialing

29 Special Issues n Hidden populations n Sample size u statistical measures (degree of confidence, variation) u “rule of thumb” F smaller sampling size, larger ratio F # of variables & attributes

30 Sample Size? n Statistical methods to estimate confidence intervals—(overhead) n Past experience (rule of thumb) n Smaller populations, larger sampling ratios n Factors: n goals of study (number of variables and type of analysis) n of populations n features of populations

31 Survey about football (soccer) market n http://www.sportfive.com/index.php?id=318 &L=1%20%282#1379 http://www.sportfive.com/index.php?id=318 &L=1%20%282#1379 http://www.sportfive.com/index.php?id=318 &L=1%20%282#1379

32 Sampling Advice for Development Project n Rural poverty project and sampling issues projectsamplingprojectsampling

33 More Issues/notions in Probability Sampling Probability More Issues/notions in Probability Sampling Probability n Assessing Equal chance of being chosen n Standard deviation n Sampling error n Sampling distribution n Central limit theorem n Confidence intervals (margin of error)

34 If time: Introduction to Standard Deviation 1 1 2 3 4 5 6 7 8 Neuman (2000: 321)

35 Calculation of Standard Deviation Neuman (2000: 321)

36 Standard Deviation Formula Neuman (2000: 321)

37 Calculation of Standard Deviation Neuman (2000: 321)

38 Interpreting Standard Deviation n amount of variation from mean n social meaning depends on exact case

39 Logic of Sampling n Use samples to make inferences about target population n Note: u Distinction between descriptive and inferential statistics u probabilistic sampling techniques needed for inferential statistics


Download ppt "Sampling. Why Sample? Some Issues: n Time, cost, accuracy n accuracy/ representativeness n Link to interesting general introduction of sampling for public."

Similar presentations


Ads by Google