research 10/20

cluster sampling

, Cluster sampling: Definition: a cluster is a sampling unit with which one or more listing units is associated. Within the cluster is the listing unit.

4 non prob sampling and charac.

, convenience: easy access 2, purposive (judgment): pre-determined 3, quota: pre-planned number 4, snowball: referrals

quick quiz

1, In the health care setting (hospital), doctors, residential medical students, medical assistants, nurses, lab technicians, social workers, dieticians, non-medical personnel all work there. Which sampling method did the researcher use in order to find a sample of 100 out of 1,000 population? __________ A: B: C: stratified sampling D: 2, In the health care setting (hospital), the researcher needs a sample of 100 professionals out of 1,000 population to understand the work stress there. The researcher first picked up six larger units - oncology, labor/delivery, ICU,UUI, emergency room, pediatric. Then, under each of the six units, each unit has a couple of sub departments. For instance, Emergency Room has family medicine sub department, orthopediatric, radiologist, etc. Lastly, we pick up sample participant from each each sub department. Which sampling method did the researcher use?___________ A: B: C: D: cluster sampling Six larger units ------- sub departments -------- professionals

systematic examples

1, alphabetic order 2, zipcode 3, $income from low to high 4, weight, from light to heavy 5, height, from short to tall 6, age, from young to old

characteristics of dots

1, dots are listed consistently or inconsistently. 2, dots are regular or irregular 3, dots are even or not even 4, dots are equally or not equally distributed.

4 types of probability sampling with description

1, simple random 2, systematic 3, stratified ; strata (stratum)=divide the whole group into subcategories according to certain characteristics. E.g. divide by gender: male (Max, Isaac, DeShawn), female (WU, Kim, Amy, Valerie, etc. ) Note: within the stratum, units are the same (homogeneous). Across the stratum, units are different (heterogeneous). 4, cluster

Probability sampling 4 types

1, simple random sampling (simple ) 2, systematic random sampling (systematic) 3, stratified random sampling (stratified) 4, cluster sampling (cluster)

stratified sampling

3, Stratified sampling (StS): Definition: Divide the population by certain characteristics into homogeneous subgroups (STRATA). E.g. By educational level: PhD students, Masters students, Bachelors students. E.g. strata by marital status: single (80%), co-habited (6%), married (7%), divorced (2%), married but separated (4%), widowed (1%) Note: STRATA is a complete list of categories for the variable of interest. Elements within each strata are homogeneous, but are heterogeneous across strata. A simple random or a systematic sample is taken from each strata relative to the proportion of that stratum to each of the others.

charac. of simple random sampling

All subsets of the frame are given an equal probability. Each element of the frame thus has an equal probability of selection. This is done by assigning a number to each unit in the sampling frame. A table of random number or lottery system is used to determine which units are to be selected.

cluster sampling characteristics

Cluster sampling characteristics: Some populations are spread out (over a state or country). Elements occurs in clumps (towns, districts)- Primary Sampling Units(PSU) (ex: schools-classroom-students ; county-hospital-patients;) You can not assume that any one clump is better or worse than another clump. Elements are hard to reach and identify. Example: state-va-veterans; state- branches/ military base - veterans; When to use it? Researchers lack a good sampling frame for a dispersed population. The cost to reach an element to sample is very high. Usually less expensive than SRS but not as accurate: Each stage in cluster sampling introduces sampling error - the more stages there are, the more error there tends to be.

systematic sampling

Definition: Starting from a random point on a sampling frame, every nth element in the frame is selected at equal intervals (sampling interval). (1), Starting point: not start from the first unit; start from a random point. (2), Interval: nth element It relies on arranging the target population according to some ordering scheme, and then selecting elements at regular intervals through that ordered list. It involves a random start and then proceeds with the selection of every kth element from then onwards. Ex: select every 10th name from the telephone directory.

simple random sampling

Definition: a sample size n is drawn from a population N in such a way that every possible element in the population has the same chance of being selected. n= sample size; how many in the sample N=population size Applicable when population is small, homogeneous and readily available. What are some ways for conducting SRS? Random number table, drawing out of a hat, random timer, etc.

In class examples of simple and systematic sampling

Ex: You have a sampling frame (list) of 10,000 people and you need a sample of 1000 for your study. How to do it? N= 10,000; n= 1,000 Simple sampling: Systematic sampling: telephone book (names are ordered alphabetically), random pick up 2nd person as the starting point, every 5 person will be picked up. 2nd, 7th, 12th, 17th, .... Q: if I start with 12nd person, can I pick the rest by every 10th? A: NO. It will go out of the range. We can pick every 5th. Look at here: 10,000/10= 1000. What if every 5th? 10,000/5 =2,000 What kind of ordering schemes? Directory; class roster;

4 non probability sampling

Four types: 1, convenience sampling 2, purposive 3, quota 4, snowball

in systematic sampling

Gaps between elements are equal and Constant There is periodicity

characteristics of stratified sampling

It involves two steps(1), define your strata, a complete list of categories, (2), within each subgroup of strata, you can use either simple or systematic to identify your final number. E.g. strata by marital status: single (80%), co-habited (6%), married (7%), divorced (2%), married but separated (4%), widowed (1%) I will select the sample n=100 by stratified sampling method. First, I have 6 strata / categories according to marital status. Second, I need to decide how many to be selected from ach category. Strata defined by different marital status: 6 subgroups. Strata by gender: male vs. female Strata by race: Caucasian, African American, Asian, Hispanic, Others. Strata by education degrees: high school or below, bachelor, master, phd or equvalent

convenience sampling

Non1: convenience sampling: Definition: it is the sampling in which one relies on available subjects. No reason tied to purposes of research. E.g. students in the class; friends; people on the street. Characteristics: Easy Small Free Quick Restrict to certain question you ask, exclude sensitive questions This approach is less scientific, hard to make inference (generalization).

quota sampling

Non3: quota sampling: Definition: Researchers select pre-plan number of subjects in specified categories (e.g. 100 men, 100 women) Focus on final sample size

another example of systematic

Population of community (zipcode) in U.S.: Total number of communities: 41,742 Sample of 1,000 communities regarding their distribution of median household income. (1), order all the median household incomes from lowest to highest. 22333, $30,000 55666, $60,000 97560, $150,000 (2), pick up your starting point 50th out of 41,000. (3), pick up the remaining sample every 40th. (50th, 90th, 130th, 170th, )

example of convenience

Q: Dr. Wu plans to study students' test anxiety and its impact on students' grade. She will include 10 students from her SWK697 class into the final sample. These 10 students are available to answer her qnairs in their convenient time. Which sampling method? _convenience____

cluster sampling example

Q: NSU N=8,000 students. A researcher plans to use cluster sampling to select 100 students for his research. How to pick up? 1, We have 5 schools/colleges: social work, educ, business, COLA, CSET. 2, We pick up 2 classrooms from each school. Total 10 classrooms are selected. 3, we pick up 10 students from each classroom. Cluster: disciplinary school ; classroom; individual student. Cluster: multiple layers.

another example of cluster sampling

Q: We have 8,000 students at NSU. 6,000 commute students, and 2,000 residential students. NSU has 5 colleges/schools (school work, education, business, college of liberal arts, college of engineering). N= 500 in the sample How to find 500 students by using cluster sampling? Colleges ---- majors ----- students Follow these steps: 1, select 5 colleges/schools 2, for each college/school, the researcher chose 5 majors. How many majors did you choose? 5*5=25 majors 3, for each major, you chose 20 students. How did you come up with 20 students? 500/25=20 students per major

example of cluster

Q: salvation army try to identify kids to participate in Christmas angel tree program. I will use cluster sampling to identify those kids. Cluster sampling: schools-classroom- kids <12 yrs old. Cluster sampling: neighborhood (zipcodes), households, kids.

another example of stratified sampling

Q: this is related to stratified sampling. NSU total N= 8,000. Some students are commute students (6,000= 75%), some are residential (2,000=25%). Q: sample n= 500. how many students in this sample should be commuters? ___________ 1, N=8,000 2, strata: commute vs. residential 3, n= 500 4, 6,000/8,000= 75%; 500*75% = 375. I will include 375 commute students in the final sample. Note: the percentage of certain category will be same in the population vs. in the sample. 75% of the population is single; 75% of the sample is single. 25% of the population is residential; 25% of the sample is also residential. How to get 25% 2,000/8,000=25%

systematic sampling short answer

Short answer Q: we have to select 5 students out of 23 in SWK697 class. How to do that with systematic sampling? 1, Choose the starting point: starting 5th student. 2, select the frequency: every fourth student will be included in the sample. A, b, c, d, e, f, g, h, i, j, k, L, m, n, o, p, q, r, s, t, u, V, w, Q: who will be included in my final sample? E, I, m, q, u Note: you can not go above the range of your population.

stratified vs cluster

Stratified: focus on the difference among the population . Different subgroups are people. Cluster: focus on levels to reach people. Q: hospital personnel can be divided into doctors, nurses, social workers, technicians, helpers. Answer:__________

in random sampling

The gap, or period between successive elements is random, uneven, has no particular pattern.

example of purpose or judgment sampling

The researcher plans to study the effect of drug additives on homelessness. He purposely selected the substance abuse clinics in town to find the sample participants. Which sampling method? __purposive sampling_______

when to use stratified sampling and example

When to use this type of sampling? When a stratum of interest is a small percentage of a population and random process could miss the stratum by chance. When enough is known about the population that it can be easily broken into sub-groups or strata. Example: find out the life expectancy for American people. (Strata:______________) Q: how to find the strata? (1)By gender by birth, female & male; (2) by race/ethnicity, (3) by age, 0-5, 6-11,

example of stratified sampling

e.g. total # students at NSU is 8000. Total # single students are 8000*80%=6400. You plan to have a sample n=1000 to start your survey. How many single students should be included? _______________________ 1, Total N=8,000: 2, By strata: single (? ), married (?), co-habited, widowed, and other kinds of arrangement 3, sample n= 1,000 4, how many single students will be included out of 1,000? 1,000*80%= 800

purpose sampling or judgment sample

purposive sampling (or Judgment Sample): Definition: select a sample based on one's knowledge of a population or drawing a sample with some predetermined characteristics in mind. Subjects selected for a good reason tied to purposes of research. The sample depends primarily on researchers' subjective judgment. E.g. students select dissertation committee. Once you determine your list of predetermined characteristics, you finalize the sample Research interests; faculty who can work well with students; more time;

snow balling sampling

snowball sampling: Definition: this technique relies on referrals from initial subjects to generate additional subjects. It can reduce the cost. In sum: (1)non-probability is less scientific than probability sampling, more subjective, less representative. (2), non-probability is more likely used in QUAL study (small sample ); probability sampling is more likely used in QUANT study (large sample ). Example: In the social service setting, the customers come in. The worker can first direct them into appropriate department within the social service building. If the issue is not resolved, they will be directed to community agencies. Referral services were provided. If the researcher plans to pick up those clients into her final sample, she will use this referral mechanism to collect all the participants.