Data collection is any process of preparing and collecting data, for example, as part of a process improvement or similar project. The purpose of data collection is to obtain information to keep on record, to make decisions about important issues, or to pass information on to others. Data are primarily collected to provide information regarding a specific topic. Data collection usually takes place early on in an improvement project, and is often formalized through a data collection plan which often contains the following activity: 1. Pre collection activity — agree on goals, target data, definitions, methods 2. Collection — data collection
3. Present Findings — usually involves some form of sorting analysis and/or presentation. There are two broad categories of data collection in research: 1. Probability Sampling
2. Non-probability sampling
* Probability sampling is a random method of selection in a targeted population. To conduct randomized samples, you need to make sure everyone in the population is given an equal chance to be chosen. Simple Random Sampling
* The simplest sampling technique is the simple random sampling, which is a lottery method of randomly picking from the targeted population. For instance, if a thesis is about malnourished students in a school, your sample size is 50 and there are 200 malnourished students, put all 200 names in a hat and pick out 50.
Stratified Random Sampling
* Stratified or proportional sampling aims to find a population for the entire population and for subgroups within the population. Taking the example on the previous technique, in the population of 200, there are 100 fifth-grade students, 50 second-grade students and 50 third-grade students. Since the sample size is 50 -- 25 percent of the population -- you need to take 25 percent from each of the three grade levels. As a result, you would have 25 fifth-graders and 12.5 second-graders and 12.5 third-graders. After determining the number of samples per grade, proceed to the lottery method.
Systematic Random Sampling
* Given that the total population is 100 and you need a sample of 20, divide the population with the sample size -- 100/20 = 5. Since the product is 5, choose an integer between 1 and 5. Let's take the integer as 2. Divide the total population (100) by the integer (2) and you get 50. Create a list of the names of the subjects alphabetically in two columns, because the integer is 2. Write the names vertically. Following the chosen integer once more, make two counts continuously. Every row that lands on the count of two is included in the sample. Cluster Random Sampling
* Stratified and systematic random sampling becomes a problem for large sample sizes, such as an entire country. Cluster random sampling limits the population by creating subgroups within the population. For example, the states on the West Coast could be one group and states in the east could be another. Multi-Stage Sampling
* Most research requires a more complex sampling method, and applying a combination of simple, stratified, systematic and cluster random samplings called multi-stage sampling addresses this need. Non-probability Sampling
* Non-probability sampling does not involve random sampling. Although researchers consider random sampling to be more reliable, it is not always the sensible or practical technique to use. Non-probability sample techniques are accidental sampling or purposive sampling. Accidental Sampling
* An example of accidental sampling is the news media interviewing people on the street. This technique is used to get a quick public opinion. Another example of accidental sampling is when college professors use students or medical researchers use available clients as a matter of convenience. These types of sampling do not represent the population as a whole. Purposive Sampling
* In this method, the researcher chooses the sample on his or her own because there...
Please join StudyMode to read the full document