Capstone assignment 7

300 MEDSURG Nursing—September/October 2009—Vol. 18/No. 5

Limitations. Next, authors should discuss the limitations of the part i c- ular study. This is usually a brief but honest appraisal of both innate p roblems of the design chosen for the study and any problems that o c c u rred in the pro g ress of the s t u d y. Limitations should not be overstated but written in a bal- anced manner. In this issue, T h o rnlow (2009) clearly outlined the specific limitations of the pub- licly available database she used to examine the rates of patient safety incidents and adverse events expe- rienced by hospitalized older pa- t i e n t s .

Limitations are based on the i n t e rnal and external validity or evaluative criteria of the study (Munhall, 2007). These criteria usually are standard for the type of re s e a rch conducted. For example, limitations of a qualitative study should not be based on criteria related to quantitative re s e a rc h . As another example, investigators in interventional re s e a rch need to a d d ress issues of the fidelity (con- sistency) of their interv e n t i o n . Mixed methods also have some unique characteristics related to whether the results were or were not integrated (Creswell & Tashakkori, 2007).

Implications. After discussing the limitations, authors usually write the implications and re l e- vance of their study for practice, p o l i c y, and future re s e a rch. Impli- cations for practice should be por- trayed cautiously and should be justified clearly by the data. For example, if the results are unclear in an intervention study, the inter- vention should not be re c o m m e n d-

The Discussion Section of a Research Report

s u p p o rted and which were not (American Psychological Associa- tion, 2009). If the study was descriptive, they should summa- rize the major findings.

Explanation of the findings related to literature and theory. Next, authors should consider why they obtained the re s u l t s they did. We re there confounding variables that might explain con- flicting or negative results? It is p e rmissible to speculate to some d e g ree as long as it is clear that it is speculation (Skelton & Edward s , 2000). Authors should compare their results to the results of other studies, discussing how they are similar to and dissimilar from ear- lier articles. If alternative explana- tions exist for the findings, these also should be considered and not i g n o red (Hess, 2004). In addition, ways in which this study expands knowledge of the topic for nursing or health care are outlined. This p a rt of the discussion re q u i re s authors to re read the literature with their findings in mind. They also may have to explore new a reas of the literature if there were unintended results; this is tru e especially for qualitative studies.

If the study was based on a p a rticular theory or framework, the authors also should re l a t e their findings to that theory. Did their findings support the theory ? Did the findings only support part of the theory? If the findings sup- p o rted using the theory in a new situation or with a new group, this also should be explained. Finally, possible revisions to the theory suggested by the findings can be a rgued.

F or readers, one of the most important sections of any

re s e a rch re p o rt is the discussion section because this is where they l e a rn the meaning and significance of the results. As an instru c t o r, I often find the discussion section is d i fficult for students and first-time authors to write. In this “Researc h Roundtable,” I will outline the pur- pose and usual components of a discussion section.

P u r p o s e

The discussion section explains how and why the authors obtained specific results. Authors should explain the outcomes and interpre t the findings for clinicians and other re s e a rchers (Naylor & Munoz- Vi v e ros, 2005). This involves not simply repeating the results, but helping readers to understand the i m p o rtance and contribution of the findings to the larger literature on the particular topic (Neill, 2007).

Components of a Discussion S e c t i o n

S u m m a ry of findings. In the first paragraph or two of the discussion section, authors generally summa- rize the major findings of the study. This summary should use straight- f o rw a rd, non-technical language understandable to all readers. If the study tested hypotheses or an i n t e rvention, the authors should summarize which hypotheses were

Lynne M. Connelly, PhD, RN, is an Assistant Professor, University of Kansas, School of Nursing, and Clinical Nurse Researc h e r, University of Kansas Hospital, Kansas City, KS. She is R e s e a rch Editor for MEDSURG Nursing.

Lynne M. Connelly

MEDSURG Nursing—September/October 2009—Vol. 18/No. 5 301

ed for all patients. Instead, a dis- cussion should ensue of implica- tions for future re s e a rch, such as how a study might be better con- s t ructed or what other variables may be included in future work. In addition, authors should be clear about patient or pro f e s s i o n a l g roups for which results are appli- cable (Hess, 2004). Other implica- tions for the next steps in re s e a rc h- ing this topic should be outlined. For example, is it time to move f rom descriptive re s e a rch to inter- ventional re s e a rch? Thorn l o w (2009) provided a clearly written section on the implications of her s t u d y, and related findings and nursing interventions from studies that have been successful.

Final Thoughts

Writing a clear, useful discus- sion section takes thought and con- siderable work. If it is not well writ-

ten, readers and reviewers may not have a clear understanding of the significance of study findings. Unclear writing even may pre v e n t the manuscript from being pub- lished. A discussion should avoid o v e r- i n t e r p retation of the re s u l t s , u n w a rranted speculation or infla- tion of the results, application of results to very diff e rent popula- tions, and tangential issues not related directly to the study (Cetin & Hackam, 2005; Hess, 2004). One suggestion is that authors always have their manuscript reviewed by a well-published colleague who can p rovide a solid critique and sugges- tions for improvements before sub- mitting the paper to a journal.

R e fe r e n c e s A m e rican Psychological Association. ( 2 0 0 9 ) .

P u blication manual of the Ameri c a n Psychological Association (6th ed.). Washington, DC: Au t h o r.

Cetin, S., & Hackam, D. J. ( 2 0 0 5 ) . An approach to the writing of a scientific manu s c ri p t . J o u rnal of Surgical Research, 128, 165- 1 6 7 .

C r e swell, J. W., & Ta s h a k ko ri, A. ( 2 0 0 7 ) . D eveloping publ i s h a ble mixed methods m a nu s c ri p t s. J o u rnal of Mixed Methods Research, 1(2), 107-111.

H e s s, D. R . ( 2 0 0 4 ) . H ow to write an effe c t i ve d i s c u s s i o n . R e s p i ra t o ry Care, 49( 1 0 ) , 1 2 3 8 - 1 2 4 1 .

Munhall, P. L . ( 2 0 0 7 ) . Nursing research: A qual - i t a t i ve perspective (4th ed.). B o s t o n : Jones & Bart l e t t .

N ay l o r, W., & Munoz-Vive r o s, C. ( 2 0 0 5 ) . T h e a rt of scientific wri t i n g : H ow to get yo u r research published. The Journal of C o n t e m p o ra ry Dental Pra c t i c e, 6, 164- 1 8 0 .

Neill, U. ( 2 0 0 7 ) . H ow to write a scientific mas- t e rp i e c e. The Journal of Clinical I nvestigation, 117, 3599-3602.

S kelton, J.R., & Edwa r d s, S. J. L . ( 2 0 0 0 ) . T h e function of the discussion section in academic wri t i n g . B M J, 320, 1269- 1 2 7 0 .

T h o rn l ow, D. K . ( 2 0 0 9 ) . Increased risk fo r patient safety incidents in hospitalize d older adults. MEDSURG Nursing, 18( 5 ) , 2 8 7 - 2 9 1 .