Nursing research article critiques-made easy!

Nursing research article critiques-made easy!

Vitale, Anne

OVERVIEW:

Let’s face the reality together. Whether or not a professional nurse can use information from a published nursing research article depends on our ability to understand the report and its conclusions. Some nurse’s feel immediately defeated when asked to read research reports mainly because of being intimidated by statistics, Nonsense! Very little of a research plan is statistical. If you can understand the rest of the report, you get help with statistical interpretation. Other nurses feel they have tried to read a research studies in the past, and couldn’t understand it. Remember, just don’t decide you can’t understand nursing research until you’ve evaluated or done your own critique of a report. Let’s get started….

HOW DO YOU DECIDE IF A PRIMARY REFERENCE IS WORTH READING?

An analytical approach to any literature review implies purposeful reading. You want to quickly elicit if the article’s findings are useable for you. Brink and Wood (1994) report reading the article will be useful and interesting to you if three criterion are met: (a) usability; (b) completeness, and (c) consistency. Before reading the article all the way through, scan the conclusions and recommendations. Can you use the findings? If not, you may not have enough interest to critically analyze the full report (Brink & Wood, 1994).

Next, apply the second criterion of completeness. Scan the entire article for key terms: (a) problem; (b) framework; (c) purpose; (d) design; (e) sample; (f) methods; (g) findings; (h) analysis; (i) protection of human rights (confidentiality), and (j) conclusions. These key terms should be in approximately this order. On your copy, underline each of the major topic headings for an easy reference guide to all the steps in the research plan. Sometimes, you have to label the content yourself, and attempt to jot your own key terms in the margin. You now have a basic outline of the report’s content (Brink & Wood, 1994).

Apply the third criterion of consistency. Identify if the study has any gaps or areas left out. If you think you missed areas in your original scan, rescan the article. Are there too many gaps? Remember, look for a comprehensive, logical progression of the material in the report, similar to the key terms above (Brink & Wood, 1994; Burnes & Grove, 1997).

If the answers are yes to the criterion of usability, completeness and consistency, the article is worth reading!

OUTLINE OF BASIC

CRITIQUE STEPS (understanding quantitative research):

Brink and Wood (1994), and Burnes and Groves (1997), similarly describe the steps for quantitative nursing research process and methods. The authors include not only the basic steps to conduct nursing studies, but also translating the studies into articles for publication. According to Burnes & Groves (1997), and Hamric and Spross (1992), the quantitative research method is used to describe or gain more information, test relationships, examine cause and effect relationships, and uses numeric data to “answer” problems. The following is a basic outline of what is included in nursing research study, and what you should critique a study for:

1. Purpose

a. Usually found in introduction or problem statement.

b. Might be stated as a main research question or hypothesis for the study.

c. Main focus may not be clearly labeled requiring reader to synthesize the purpose from this section.

2. Sample

a. Look for representativeness of sample.

1. Representativeness-Subjects are randomly selected from the target population.

2. Target population-Population from which the sample is chosen and study findings are generalized to. Example: All women ages 65 to 90 with a diagnosis of acute MI.

3. Sample size-The sample size should be as large as possible. Sampling error decreases as sample size increases (general rule).

b. Random Vs. non-random sample

1. Convenience, strictly voluntary sampling selection (usually non-random) may have bias in representativeness.

2. Bias in sampling selection means those chosen to participate may differ from those not chosen. A randomized sample reduces bias in sampling selection.

NOTE: Many nursing primary references are based on non-random convenience samples! Bias may be reduced through certain data analysis techniques, and addressed in design limittions.

3. Methods

a. Data collection – Procedure should be clearly defined (based upon problem and sample).

1. Description of data collection time frames, where data collected, participant’s permission procedure, and confidentiality/anonymity assurance.

2. How questionnaires, scales and/or interviews were utilized in the study.

NOTE: Method of data collection can also have bias. Use of interview technique to gather data has the least bias because the investigator has less influence on the participant’s answers.

b. Study Designs

1. Permits the examination of the study’s research variables. Variables are qualities, or characteristics of persons, things, or situations under study.

2. Descriptive-Gain more information about characteristics of a group. Example: pilot or exploratory.

3. Correlational-Examines relationships between variables in a group.

4. Quasi-experimental/ Experimental-Examines causality, explains relationships, uses control and experimental groups.

NOTE: Experimental designs are the most scientific!

c. Measurements/Instruments

1. Description of the data collection instruments, scales, questionnaires. Example: # questions, scoring range, etc.

2. Inclusion of reliability and validity of instruments.

a. Reliability-Measurement of how consistently similar results are obtained every time the scale/instrument is used.

b. Validity-Measurement of how accurately the instrument reflects some of the variables in the study (characteristics under study).

3. Reliability and validity are important because the study’s results should never be influenced by instrument/scale error.

4. Data Analysis:

a. Summarizes and describes the data in a logical, understandable format from research variables capable of being quantified (converted to numbers).

1. Descriptive statistics-Clearly and understandably describes the sample mostly using frequency distribution, mean, median, mode, range, % and others.

2. Inferential statistics-Tests the research questions or hypothesis using T-Tests, ANOVA, multiple regressions, etc. When the research plan hypothesizes relationships between variables, it is necessary to clarify expected variable relationships using independent and dependent variables.

a. Dependent variable-Response, behavior, or outcome the researcher wants to explain.

b. Independent variable-The treatment or experimental variable manipulated by the researcher to create an effect on the dependent variable.

NOTE: One easy way of differentiating between these variables is to remember that independent means standing alone, and dependent means relying on something. Don’t let statistics scare you! If you are a novice at article critique, work on being able to find the sample adequately represented in an included table, chart or graph. Take small steps!

5. Findings/Nursing Implications/ Conclusions:

a. The findings and conclusions should be relevant, useable, and generated from the study.

b. Suspect bias if questionable generalizability, sample defect, or weak design.

NOTE: Don’t expect perfection! Every report might offer something useful, ideas you can use, improve, or develop!

“POLISHING” THE CRITIQUE

Remember that all research is subject to a critique. But until you, the reader have critiqued the report, you have no way of knowing if the study is based on good, systematic research principles. Follow the above outline next time you read a nursing research article. Jot down the outline headings (purpose, sample, methods, data analysis, and nursing implications). And find the content from the article you chose to read. For the first couple of studies, critique the article backwards and forwards. This is time well spent because you will have an opinion on the value of the nursing study by the end of the report. Now let’s see how objective that opinion is!

“TEST THE STUDY”

Brink and Wood (1994) include a simple assessment to assist you in determining the adequacy of the nursing research you have read. The more adequate the study, the more you can learn from the research conducted! After you have thoroughly read your chosen article, apply “test the study” (Brink & Wood, 1994:

“Test the Study” involves giving the article a study rating:

A. Defective study-Lack clarity significance.

B. Substandard study-Lacks completeness.

C. Adequate Study-Covers basic research objectives.

Follow these simple steps:

1. Get out paper and pencil.

2. List each of the major portions-problems/purpose, sample, methods, analysis, findings and conclusions.

3. Across the top of the paper list categories: defective, substandard and adequate.

4. Check off which heading each section of the article falls.

5. Add up checks in defective column and multiply x1.

6. Add up checks in substandard column and multiply x2.

7. Add up checks in adequate column and multiply x3.

8. Total scores:

> 12 points-Adequate study.

8-11 points-Substandard study.

Implications:

1. An adequate statement of purpose clearly describes what the study is about.

2. An adequate sample is clearly related to the population being examined and has representative ness.

3. An adequate report on methods will tell you what, why and how the study was done in sufficient detail.

4. An adequate analysis is comprehensible, responsive to the data, and congruent with all presented material.

5. An adequate conclusion has a sense of finality and closure and is derived from the problem and reflects the findings (Brink & Wood, 1994).

CHECK YOUR SCORE AGAINST THE FEELING YOU HAD AFTER YOU THOROUGHLY READ THE ARTICLE-DO THEY AGREE, OR IS THE SCORE TOTALLY INCONSISTENT WITH YOUR FEELINGS?

Always use your objective scoring until after practice, your feelings and the scores agree consistently! You can critique nursing research articles!.

Happy reading!

References

Burns, N. & Grove, S. (1997). The practice of nursing research: (Critique, and utilization. (3rd ed) Philadelphia: W.B. Saunders Co.

Brink, P. & Wood, M. (1994). Basic steps in planning nursing research from question to proposal (4th ed). Boston: Jones and Bartlett Publishers.

Hamric, A. & Spross, J (1992). The clinical nurse specialist in theory andpractice. Philadelphia: W.B. Saunders Co.

Watson, J. (1085). Nursing: The philosophy and science of caring. Colorado Associated University Press.

By Anne Vitale, MSN, APRN, BC

Director of Nursing Research At Community Medical Center, an Affiliate of the Saint Barbabas Health Care System

Copyright New Jersey State Nurse’s Association Dec 2003

Provided by ProQuest Information and Learning Company. All rights Reserved