Literature review, Questionnaire preparation, Data collection, Data analysis in SPSS, Report writing

© Quantitative Research for ITTI Lecture 3 Population and Sampling © Sampling is the process of selecting cases to participate in your study Compose frame Define population Solve frame problems Choose census or sample Design sampling procedure Determine sample size Validate sample Choose type of sample © First, define the population, the collection of elements about which inferences are to be made Elements could be institutions, persons, groups Specify elements, place, time © Second, obtain the sampling frame, a list of all elements in the population Sampling frame: List of names and addresses of X’s Tokyo -bound customers ______________________ _______________________ _______________________ ____________________ ______________________ ______________________ ______________________ ______________________ ______________________ ______________________ ______________________ ______________________ Population: Tourists visiting Tokyo with Tour Operator X in May 2013 © Third, identify and solve any problems with the frame Sampling frame: List of names and addresses of X’s Tokyo -bound customers Alinda Kokkinou John Smith, Sr. Bob Jones Jeroen Nawijn Bob Jones Stephen Mattes Jana Machova John Smith, Jr. Alien names: not on trip… Sometimes you have duplication… Missing: Bob’s brother Frank © Fourth, decide if you will use a census or sample Would you have enough resources (time, money, people) to study the entire population? How big is the population? (up to 1000, a census may be worthwhile) SAMPLE CENSUS © Fifth, choose a probability and a non-probability sample, depending on need to generalize Unequal (or unknown) chance of being selected Everyone has the same (or a known) probability of being selected PROBABILITY NON-PROBABILITY can can not © Sixth, determine the sample size you need according to your planned analysis…… favorite SM fb tw tu inst snpcht total business tourist total N = 250 Base the plan on your largest possible cross-table You want minimum 25 cases per cell 2 rows 5 columns 25 250 x x = © ……or, according to a formula Formula: prec.=z (p x q)/n Precision: 3% (0,03) Confidence level: 95% (z = 1,96) Variability: P = 0,5 (q=1-p) 0,03 = 1,96 (0,5 x 0,5)/n 0,03² = 1,96 ² x 0,25n 0,0009 = 0,9604n n = 1067 Depends on heterogeneity of population… level of precision… level of confidence… …not on population size. www.surveysystem.com/sscalc.htm (or google ‘sample size calculator’) © 1 © © © © © © © © © © © © © © © © Seventh, plan in terms of collecting gross sample size so your net sample is large enough Refuses to participate Add expected non- response to calculated sample size You messed up their address Dies during study © Eight, decide on a specific type of sample, either a probability sample…… Random sample––every individual has equal chance of being selected Systematic sample––every Nth individual 1st 7th 3rd 5th Stratified sample–– random sample within each subgroup © ……or a non-probability sample , which is not generalizable but often easier Convenience/opportunity sample Select whoever is available Snowball sample Ask individuals to suggest other potential participants Quota sample Sample at least minimum from each subgroup © Once you select your sample, the way you contact them can greatly impact the response you get Post? Email?

In -person? © Finally, validate your sample = Does sample distribution on prominent variables resemble the population distribution? [Have to know something about population] % % % Choose a variable that’s known in the population and check if your sample is not too different © Read before workshop:

 Bryman book 4 th ed.: pp. 330-339 Please bring your book to the workshop, a print copy of the Course Outline and a print copy of your survey. Read before next lecture:  Bryman book 4 th ed.: pp. 339 -350