Background - Cengage



Background

How representative data are collected is fundamental to criminal justice research. There are three sources of data available to criminal justice researchers:

1. asking questions

2. direct observations

3. examining existing data

As a research or evaluation project is designed, choices must be made about measurement and data collection. Sometimes, multiple measures of some concept need to be incorporated into a single research design.

Decisions about data collection must take into account a number of general issues:

( Researchers must be attentive to the validity and reliability of their measures.

( Social scientists distinguish between obtrusive and unobtrusive measurement. In obtrusive or reactive measurement, research subjects are aware that they are being studied. In unobtrusive measurement, the subjects are not aware.

( Researchers must be careful, but also creative. Because measurement is so important but often so difficult in criminal justice research, it is frequently necessary to devise creative substitutes for direct measures.

Probability sampling is a method of selection in which each member of a population has

a known chance or probability of being selected. This type of sampling technique helps

researchers generalize from observed cases to unobserved ones. Sampling bias may be

conscious or unconscious. A basic principle of probability sampling is that a sample will be representative of the population from which it is selected if all members of the population have an equal chance of being selected in the sample. Samples having this quality are often labeled EPSEM samples (equal probability of selection method).

Probability theory permits inferences about how sampled data are distributed around the value found in a larger population. There are four important terms that relate to probability sampling:

( sample element – that unit about which information is collected and that provides the basis of analysis

( population – the theoretically specified grouping of study elements

( population parameter – the summary description of a given variable in a population

( sample statistic – the summary description of a given variable in the sample (used to make estimates of population parameters)

The key to the process of probability sampling is random selection: This procedure

serves as a check on conscious or unconscious bias; and, more importantly, random selection offers access to the body of probability theory, which provides the basis for estimates of population parameters and estimates of error. The sampling distribution is defined as the range of simple statistics that would be obtained if many samples were selected. In order to draw an actual sample, a sampling frame is needed – a list of elements in the population. In probability theory, the standard error is a valuable piece of information because it indicates how closely the sample estimates will be distributed around the population parameter. The logic of confidence levels and confidence intervals provides the basis for specifying that a certain percentage of a fictitious large number of samples will produce estimates that fall within one or more standard errors of the parameter. This logic also provides the basis for determining the appropriate sample size for a study.

The correspondence between a target population and sampling frames affects the generalizability of samples. Once again, a sampling frame is the list or quasi-list of elements from which a probability sample is selected. Sampling frames may be considered as operational definitions of a study population.

Different types of sampling designs can be used alone or in combination for different research purposes. Simple random sampling forms the basis of probability theory and the statistical tools we use to estimate population parameters, standard error, and confidence intervals. Simple random sampling is seldom used in practice and researchers usually use systematic sampling. This method involves researchers choosing all elements in a list for inclusion in he sample. The one danger of systematic sampling is that a periodic arrangement of elements in the list of elements can make this technique unwise; this arrangement is usually called periodicity. Stratified sampling represents a possible modification that allows for obtaining a greater degree of representativeness and decreasing the probable sampling error. Disproportionate stratified sampling involves purposively producing samples that are not representative of a population on some variable. Cluster sampling may be used when it is either impossible or impractical to compile an exhaustive list of the elements that compose the target population. Population elements or aggregations of those elements are referred to as sampling units. Multistage cluster sampling can be combined with stratification in order to improve the sample being studied.

The text utilizes the National Crime Victimization Survey (NCVS) and the British Crime Survey (BCS) to illustrate different ways of designing samples to achieve desired results.

In many research applications, nonprobability sampling procedures are necessary or advantageous. In nonprobability sampling, the likelihood that any given element will be selected is not known. Four types of nonprobability samples are examined in the text. Occasionally, it may be appropriate to select a sample on the basis of our own knowledge of the population, its elements, and the nature of our research aims – in short, based on our judgment and the purpose of the study. Such a sample is called a purposive sample. Quota sampling begins with a matrix that describes the characteristics of the target population we wish to represent. Quota and purposive sampling may be combined to produce samples that are intuitively, if not statistically, representative. Relying on available subjects – that is, stopping people at a street corner or some other location – is sometimes called convenience sampling. Most commonly used in field observation studies or specialized interviewing, snowball sampling begins by identifying a single or small number of subjects and then asks that subject to identify others like him or her who might be willing to participate in a study.

Like other elements of criminal justice research, sampling plans must be adapted to specific research applications. When it is important to make estimates of the accuracy of samples, and when suitable sampling frames are possible, probability sampling techniques are used. When no reasonable sampling frame is available, and a probability sample cannot be drawn, estimates about sample accuracy cannot be made, but in these instances, a variety of approaches can be utilized for drawing nonprobability samples.

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download