Share

Design Decisions in Research

4. The Design and Planning Phase

Validity

In quantitative research, validity is a quality criterion that indicates the degree of accuracy of study conclusions (Polit & Beck, 2004). It is important to recognize that numerous variables, in addition to the variables examined in a study, may be influencing the results and thereby posing threats to the validity of conclusions (Pedhazur & Schmelkin, 1991). Anticipating and controlling threats to validity requires careful consideration when designing a quantitative study.

Types of validity include: internal validity, external validity, construct validity, and statistical conclusion validity.

Table 3

Validity Standards in Quantitative Research

Definition
Internal validity Extent to which the effects detected in the study are a true reflection of reality rather than the result of some other extraneous factor
External validity Extent to which the study findings can be generalized beyond the sample of the study
Construct validity The fit between what an instrument is intended to measure and what the instrument actually measures
Statistical conclusion validity Whether conclusions about relationships or differences drawn from statistical analysis are an accurate reflection of reality

Table previously published in: Whittemore, R. & Melkus, G. (2008). Designing a research study. The Diabetes Educator, 34, 201-216.

Internal validity, or the degree to which study results are true and can be attributed to the variables measured, is the most important consideration in designing a quantitative study. There are many research strategies to enhance internal validity, including (Pedhazur & Schmelkin, 1991):

  • Random selection of study participants;
  • Random assignment of participants to groups;
  • Inclusion of a control group;
  • Modifying sampling criteria; or
  • Using statistical analyses to control confounding variables.

Example 2


For example, in a study in which depressive symptoms may influence the results, the investigator may want to exclude participants who are depressed or the investigator may want to include participants who are depressed and statistically control for this effect on the outcome variable. Both of these strategies would enhance the internal validity of the study.

Pedhazur, E. J., & Schmelkin, L. P. (1991). Measurement, design, and analysis: An integrated approach. Hillsdale, NJ: Lawrence Erlbaum Associates Publications.
Polit, D. F., & Beck, C. T. (2004). Nursing research: Appraising evidence for nursing practice (7th Edition). Philadelphia: Wolters Klower/Lippincott Williams & Wilkins.