Home Up Library CD ROM Computer Logic Introduction Literature Procedures Sampling

Research Design in Occupational Education
Copyright 1997. James P. Key. Oklahoma State University
Except for those materials which are supplied by different departments of the University
(ex. IRB, Thesis Handbook) and references used by permission.




The information or data gathered to achieve the objectives, answer the research questions, or test the hypotheses is really the basis of the entire procedures chapter. This needed information determines the necessary design, the population and samples required, the situation involved, and the methods and tools needed. You must ask "Will this information adequately achieve the objectives, answer the questions, or test the hypotheses and enable me to achieve the purpose of the study?" If the answer is yes, then it should be relatively easy to establish the design and procedures for the study. If the answer is no, then you must determine what information you do need before going further.


Research Design

Research design refers to the systematic steps set up to accomplish the purpose of the study. The purpose determines the steps and the order in which they are arranged and conducted.

For example, if the purpose is to describe a situation, then the steps of a survey or case study would be appropriate and the specific objectives or research questions would help determine the order of the steps. However, if the purpose is to determine the effectiveness of a particular teaching method as compared to another method, the steps of a quasi-experimental design would be appropriate and the specific hypotheses to be tested would help determine which design would be most efficient.

At the beginning of the procedures chapter, the research design should be thoroughly described to establish the setting for the choice and description of the tools and methods needed to fit into this design to achieve the purpose.


Population vs. Sample

To answer the question of Who? and determine whether to utilize a population or sample in obtaining the necessary information, several ideas need to be considered. First, to what group of subjects or objects do you wish to generalize the findings? The answer will begin to establish the limits of the population.

After establishing the limits of definitions of the population, the size needs to be evaluated to determine how many subjects or objects it includes. If you are conducting a descriptive survey, can you describe the entire population, or due to time and resource constraints should you try to obtain a representative sample? If you are conducting an experimental study, the most common approach is to use samples.

In determining the sample size, a rule of thumb is that the larger the sample the more accurate the findings will be. Statistics books have formulas for determining samples sizes based on population heterogeneity and accuracy desired. After determining whether to use a population or sample, the population and/or sample need to be explicitly described in terms of geographical limits, types of groups, special identifying characteristics, and any other useful identifications.



All important details about when and where the study will be accomplished need to be included in a description of the situation of the study. If the study is longitudinal, it may extend over a lengthy period of time, perhaps several years; whereas, a normal study will extend over a relatively short period of time, possibly extended up to a year. The geographical limits of the area for the study should be clearly described. This also limits the area to which the findings will apply. If the population characteristics vary according to geographical regions, it may be necessary to use a stratified sampling technique, stratifying according to geographical regions.



The description of the data gathering instruments, such as questionnaires, interview schedules, tests, inventories, etc., should be adequately detailed to provide the source of the instrument, any special characteristics (including its validity and reliability), any special restrictions or limitations, and its most appropriate use. Standardized instruments usually have established validity and reliability either from a history of use or from actual tests, while researcher-designed instruments usually do not have established validity and reliability. The researcher-designed instruments may need to be validated by a panel of experts and should be pilot-tested where possible.

The information sought and population will determine the sampling methods. If inferences are to be made, a representative sample will be sought, most likely through random sampling. If the population varies according to geographical location or by special groupings (ethnic groups), then stratified random sampling may be needed. If these locations or groups vary a great deal in size, then proportional samplings within the strata may be necessary. If the population is quite widespread and cost and other limitations prevent sampling individuals within the populations, then clusters such as schools or classes within schools may be used with the randomization of clusters. If the information about the population is so adequate that a representative sample can be logically chosen, then perhaps a purposive sample might be used. However, sampling methods should be used only when measuring the characteristics of the population is impossible due to cost, time, or other resource constraint. A thorough, detailed explanation of the sampling techniques is necessary to establish the validity of the information and findings.

The statistical methods used will be largely determined by the information sought. If descriptive information is sought, then descriptive statistics, such as counts, percentages, means, and ranges, may be the most adequate statistical tool. However, if inferences from a sample to a population are desired, then inferential statistics and descriptive information will be needed. Whether comparisons or relationships are sought, the measurement level (nominal, ordinal, interval) and the assumptions to be met (normality of population, equal variances, etc.) will determine the type of inferential tool to use. This information and statistical formulas, sources, and justifications should be clearly explained in the procedures chapter.



To assess your ability to accomplish the objectives, develop the components of the procedures chapter following these guidelines. Consider and include the following criteria.

Type of Design

Population and/or Sample

Sampling Techniques

Instrument Description

Instrument Development

Instrument Reliability and Validity

Procedures for Gathering Data, including Confidentiality

Data Analysis Techniques

Statistical and Mathematical Procedures



Home Up Library CD ROM Computer Logic Introduction Literature Procedures Sampling