Formulating research questions and hypotheses in healthcare research

¶ … motivation to learn, followed by an educated and informed research question. After the research question has been narrowed down, the researcher conceptualizes how the question may be answered feasibly, within the constraints of time and budget. Such constraints can help guide the research questions into a workable hypothesis that is testable, as well as being relevant and meaningful to healthcare stakeholders. From here, the researcher can glean a cogent hypothesis.

Then begins the process of data collection, which is also determined by the research question ("Overview of the Research Process," n.d.). Data collection methods relevant to healthcare fields include everything from surveys and interviews to quantitative measures such as number of emergency room visits or amount of money spent on specific treatments.

Once raw data is collected, it is then analyzed using the help of statistical analysis software. The data analysis demonstrates whether the hypothesis was proven true or not. The researchers then analyze the results of the research using a theoretical framework they identify, to show possible reasons for the proven hypothesis or the failure to prove the hypothesis. Finally, the research process concludes with an honest assessment of the credibility of the study, and how the results of the research might contribute to a better understanding of healthcare policy or procedure.

Research Question v. Hypothesis

A research question usually arises "out of a perceived knowledge deficit within a subject area or field of study," (Farrugia, Petrisor, Farrohkyar & Bhandari, 2010, p. 278). The research questions usually precede the construction of a hypothesis, which shows that the researcher has investigated how other researchers have approached similar issues to identify the specific gap in the literature. When the specific gap in the literature has been identified, the researcher can develop a hypothesis that is testable. A hypothesis usually entails framing a specific relationship between variables, and is much more specific than a research question. The question remains relevant as a framework for the research, and for helping place the hypothesis and research results into a theoretical context.

Types of Data Collection Tools Used in Health Care Research

In healthcare research, data collection tools may include interviews and surveys for qualitative assessments, or they may include accessing quantifiable data such as the number of patients, discharges, incidents and prevalence rates, or budgetary measures. The types of data collection tools will vary depend on the research question and the variables being measured. For example, some research questions will be related to the effectiveness of a specific intervention, which would require quantitative assessments based on specific measures of success.

Types of Sampling Methodologies

There are a plethora of sampling methodologies, each with its strengths and weaknesses. Random sampling is favored in research but is not always feasible. Researchers usually rely on interval sampling, in which a population from a specific community of healthcare center is randomly sampled. This sample is then used to represent the total population. Clustered and stratified sampling are other methods that may be used in healthcare. Stratified entails selecting a sample population from disparate geographic regions to maximize diversity of the sample, whereas clustered sampling involves sampling from regions sharing similar features such as rural healthcare centers versus urban ones.

Types of Statistical Analyses

The types of statistical analyses used in healthcare research vary in terms of complexity, and depend on the nature of the research hypothesis. Qualitative research uses no statistical analyses, and descriptive data requires only the simplest measures such as mean, median, and mode. More complex statistical tests include tests for correlation, such as the Pearson correlation or the Chi-Square, which reveal the strength of the relationship between two variables. Comparing the means of larger population groups requires tests like the analysis of variance, or independent T-test, which tests the differences between two independent variables. Regression tests are used to determine whether a change in one variable predicts a change in the other variable ("Types of Statistical Tests," n.d.).