Nursing Research Nursing research involves analyzing Qualitative Data, which usually consists of narrative materials analyzed by the major styles including but not limited to Template, Editing and Immersion/crystallization analyses. One synthesized method to probe quality and integrity employs primary and secondary criteria for scrutiny during studies and during post hoc assessments of studies. Three statistical processes for use with data are categorizations of data according to Nominal, Ordinal and Interval measurements. Two ways to determine the effectiveness of statistics are Absolute Risk Reduction and Relative Risk Reduction. Nursing research also involves Statistics and Quantitative Analysis Design, which employ Inferential Statistics, assisting in the analysis of research results by drawing inferences about relationships based on observations of collected data. Multivariate Statistics is an area of statistics concerned with the collection, analysis and interpretation of several statistical variables at once. Multivariate Statistics analysis is integrated in quantitative analysis through a number of tests to compare a number of variables in complex relationships. Nursing research also involves Analysis, Synthesis, and Reporting Results, including meta-analysis, meta-synthesis, reporting the results by following a general process of: selecting the medium he/she will use to convey the information; defining the audience; developing a coordination plan including authorship, content, materials assembly and effective writing. Nursing research may also involve a research proposal that defines a research problem and proposes methods of solving the research problem, typically submitted to funding agencies or faculty advisors who accept, reject or demand modifications to the proposal.
Nursing Research
Analyzing Qualitative Data
Qualitative data, which usually consists of narrative materials (Polit & Beck, 2008, p. 507), is analyzed by three major styles. Template analysis involves development of a guide to which gathered data is applied and adjustment of the guide as more data is applied. This method tends to be interpretive rather than statistical. Editing analysis involves interpretation of data for "meaningful segments," then development of a category scheme with codes according to those segments, then a search for patterns that may connect the categories. Immersion/crystallization analysis is a reflective, subjective interpretation of data by the researcher's immersion in the data, and is rarely used in nursing research (Polit & Beck, 2008, p. 508).
Quality and integrity in qualitative studies are significant issues for the health care industry and have been addressed by numerous theorists. One synthesized method to probe quality and integrity employs primary and secondary criteria for scrutiny during studies and during post hoc assessments of studies. Primary criteria during both phases consist of probing questions regarding: credibility, authenticity, criticality and integrity. Secondary Criteria during both phases involves probing questions regarding: explicitness; vividness; creativity; thoroughness; congruence and sensitivity (Polit & Beck, 2008, p. 541).
Three statistical processes for use with data are categorizations of data according to Nominal, Ordinal and Interval measurements. Nominal measurement, the lowest and strictly quantitative measurement, assigns numbers to categorize characteristics, for example, gender, religious affiliation and marital status (Polit & Beck, 2008, pp. 556-557). Ordinal measurement sorts and ranks objects according to set criteria but does not meaningfully measure differences between different rankings, for example, the Shin study's arrangement of data according to rankings of "precontemplation, contemplation and active" criteria. Interval Measurement also uses rank-ordering and additionally employs "equivalent distance" between the ranks, allowing measurement of the differences between ranks (Polit & Beck, 2008, pp. 557-558).
Two ways to determine the effectiveness of statistics are Absolute Risk Reduction and Relative Risk Reduction. Absolute Risk Reduction essentially compares two risks by subtracting the absolute risk ("the proportion of people who experienced an undesirable outcome in each group" (Polit & Beck, 2008, p. 572)) for a treated group from the absolute risk for an untreated group. For example, David A. Alter's Therapeutic lifestyle and disease-management interventions: pushing the scientific envelope (Alter, 2007) measured the difference in global health risk between cigarette smokers who received smoking cessation intervention vs. cigarette smokers who received no such intervention. Relative Risk Reduction shows the proportion of untreated risk reduced through intervention and uses a formula whereby the Absolute Risk Reduction is divided by the absolute risk of the control group. For example, Can a statin neutralize the cardiovascular risk of unhealthy dietary choices? Measured relative cardiovascular risk reduction by dividing the absolute cardiovascular risk reduction of a control group (of 42,848 patients in 7 randomized trials) by the absolute risk of that group, thereby showing the proportion of cardiovascular risk reduced through intervention by the use of statins (Ferenczi, Asaria, Hughes, Chaturvedi, & Francis, 2010).
Statistics and Quantitative Analysis Design
Inferential statistics are based on the laws of probability and allow inferences to be drawn about a population based on a sampling of that population. Three applications for inferential statistics are: the sampling distribution of the mean; estimating parameters; testing hypotheses. The Sampling Distribution of the Mean employs an infinite number of samples from a selected population and theoretically distributes the means of those samples. Estimating Parameters consists of defining and establishing a framework for the target population from statistical samples (Polit & Beck, 2008, pp. 583-584). Finally, hypotheses are tested with objective criteria provided by data to infer whether the hypotheses are sufficiently supported by the evidence (Polit & Beck, 2008, p. 587).
Multivariate Statistics is an area of statistics concerned with the collection, analysis and interpretation of several statistical variables at once. While statistics may be artificially confined for convenience sake, health care actually involves complex relationships of variables for patients themselves, within a single health care institution, within a group of health care institutions, and within the entire health care system. Multivariate statistics observes and analyzes several of these variables at once using several types of tests for various purposes.
You’re 81% through this paper. Sign up to read the full paper.
Sign Up Now — Instant Access Already a member? Log inAlways verify citation format against your institution’s current style guide requirements.