1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November...

26
1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006
  • date post

    22-Dec-2015
  • Category

    Documents

  • view

    217
  • download

    0

Transcript of 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November...

Page 1: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

1

Qualitative and Quantitative Sampling

Social Research Methods 2117 & 6501

Fall, 2006

November 22~30, 2006

Page 2: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

2

Sampling: The process of selecting observations (抽樣 : 選擇觀察對象的過程 )(The purpose: get a representative sample)

Think about the following questions:

Q: Can we observe every one?

Q: Can we generalize our findings?

The Key: probability sampling (機率抽樣 )

Page 3: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

3

A Brief History of Sampling: related to political polling( 抽樣的發展歷史與政治民意調查有關)

The stories of American presidential elections

Literary Digest & Gallop Poll

Page 4: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

4

Two Types of Sampling Methods:

Nonprobability Sampling

vs.

Probability Sampling

Page 5: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

5

Some Terms First Element & Population

(sampling) element: the unit of analysis or case in population (元素 : 母體的構成單元或分析單位 )

Population: the abstract idea of a large group of many cases from which a sample is drawn (母體 :理論上一研究的特定個案元素的集合體 )

A target population (a study population): from which the sample is actually selected, the specific pool of cases studies (目標母體或研究母體 :實際上從中抽取樣本的元素之集合體 )

Page 6: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

6

Nonprobability Sampling

Qualitative researchers tend to use nonprobability or nonrandom sample. Qualitative researchers’ concern: relevance

Page 7: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

7

Nonprobability Sampling Haphazard, Accidental, or Convenience Sam

pling (就近取得研究對象,便利抽樣 ) Quota Sampling (定額抽樣或限額抽樣 ) Purposive or judgmental Sampling

(立意或判斷抽樣法 : 以研究目的為基礎來抽樣,通常由專家來判斷,尋找特定或一般較難尋找的對象 )

Page 8: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

8

Nonprobability Sampling

Snowball Sampling (滾雪球抽樣法 : 適用於很難找到特殊的研究對象時,或研究對象屬於一特定的團體 )

Page 9: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

9

Nonprobability Sampling Deviant Case Sampling: locate unusual, different, or

peculiar cases that are not representative of the whole (找尋極端個案 )

Sequential Sampling: similar to purposive sampling, but the difference is to gather cases until the amount of new information or diversity of cases is filled (與立意抽樣法不同的是,繼續蒐集個案直到新資訊或個案差異、多元性滿足為止 )

Theoretical Sampling: what is sampled comes from grounded theory (依紮根理論選取樣本 )

Page 10: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

10

Probability Sampling : samples selected accord with probability theory (依機率理論抽出的樣本就是機率抽樣 )

The key: a sample must contain essentially the same variations that exist in the population To control conscious and unconscious sampling bias

Page 11: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

11

Probability Theory, Sampling Distribution Element & Population & a target population

(sampling) element: the unit of analysis or case in population (元素 : 母體的構成單元或分析單位 )

Population: the abstract idea of a large group of many cases from which a sample is drawn (母體 :理論上一研究的特定個案元素的集合體 )

Sampling ratio (抽樣比 ) Sampling frame (抽樣架構 ): a list of cases in a p

opulation, or the best approximation of it (類似母體元素的列表 )

Page 12: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

12

Probability Theory A parameter (母數或參數 ) vs. a statistic (統計值 ) a parameter: any characteristic of a population (母體既有變數的特徵描述 )

A statistic: information from the sample (樣本變數的特徵描述,用來推估母體 )

A random process: Equal chance of being selected

The purpose: to select a representative sample Can calculate the sampling error, s = √[(P*Q)/

n]

Page 13: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

13

The logic of sampling

The concept of sampling distribution (抽樣分布 ) The central limit theorem

Let’s play a game!

Page 14: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

14

Sampling distribution

Page 15: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

15

Sampling distribution

Page 16: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

16

More on Sampling Distribution We can estimate the sampling error Confidence intervals (信賴區間—估計母體母數數值的幅度 ) and confidence levels (信賴水準 ) Provide the basis for determining the approximate

sample size Be careful:

Theory vs. survey conditions Tend to overestimate the precision of estimates

Page 17: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

17

Types of Probability Samples Simple Random Sampling(簡單隨機取樣 ): a

basic sampling method Can use random numbers (亂數 ) or computer Seldom used in practice

Systematic Sampling (系統抽樣 ): every kth element is chosen Usually apply a random start (隨機開始的系統抽樣 ) Sampling interval and sampling ratio (抽樣間距與抽樣比率 ) Be careful about any periodicity in the list(注意元素排列的週期性 )

Page 18: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

18

Simple Random Sampling and Systematic Sampling

Page 19: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

19

Types of Probability Sampling Stratified Sampling: to obtain greater repr

esentativeness (分層抽樣 : 減少誤差以選取更具代表性的樣本 ) Elements drawn from homogeneous subsets of a pop

ulation (從母群內同質性的次群體中取樣 ) Select variables you want to represent accurately, ex:

gender, geographical locations, social class, ethnicity Implicit stratification in systematic sampling (系統抽樣隱含的分層性) Ex: 對大學生的抽樣

Oversample a specific stratum (對母體特定階層抽取超過比率的樣本 )

Page 20: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

20

Types of Probability Sampling (Multistage) Cluster Sampling [(多階段 )集群抽樣 )]: used when a list of elements of a population not available and/or the cost to reach a sample element very high A cluster (集群 ): a unit that contains final samplin

g elements but treated temporarily as a sampling element it self

The process: randomly sample clusters, then randomly sample elements

Draw several samples in stages

Page 21: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

21

Page 22: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

22

Cluster Sampling Cluster sampling and sampling error: less expensive

but less accurate (集群抽樣成本較低,但是樣本較不正確,因其抽樣誤差較大 ) Each stage introduces sampling errors, so multistage cluste

r sample has more sampling errors than a one-stage random sample.

Cluster design: how to decide the # of clusters and the # of elements within clusters? The key: more clusters is better (多選取集群 ), why?

Within-household sampling: to select the individual within the household randomly

Stratification in multistage cluster sampling We can apply stratification techniques at each stage.

Page 23: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

23

Cluster Sampling—Probability Proportionate to Size (PPS) (隨樣本大小比例的隨機抽樣 ) Proportionate or unweighted cluster sampling:

because the size of each cluster is the same But the more common situation is for clusters

to be of different sizes. (常見的情況是集群大小不一致 )

So, we need to use probability proportionate to size (PPS) technique. The key: each element has an equal chance to be

selected into the sample

Page 24: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

24

Other Sampling Issues Random-Digit Dialing (RDD): used when the

general public is interviewed by telephone The sampling element in RDD is the phone numb

er Studying hidden populations: apply nonproba

bility and probability sampling Sample size: how large should a sample be?

A smaller population needs a bigger sampling ratio.

Consider three things: accuracy, variability, the # of different variables examined

Subgroups

Page 25: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

25

Other Sampling Issues Drawing Inferences (推論 )

Why sampling? Can draw inferences from the sample to the population.

Combining logics of sampling and measurement Validity and sampling error

Page 26: 1 Qualitative and Quantitative Sampling Social Research Methods 2117 & 6501 Fall, 2006 November 22~30, 2006.

26