Recruitment and Retention of Nursing Staff from a Nurse Manager Perspective: Analysis of Data and Interpretation of Results

Techniques Used to Analyze Data

Once all the data has been collected, results will be analyzed using IBM's Statistical Package for Social Sciences (SPSS; IBM, 2013). SPSS is a statistical program that allows for simple computations and analysis of large sets of data, such as that utilized in the current study. This package allows for the calculation of both descriptive and inferential statistics, both of which will be critical to answering the study questions described above (Burrill, n.d.). Specifically, a multiple regression analysis will be conducted to determine the impact of various factors in nursing staff retention rates. While the main hypothesis in the current study is that perceptions of upward mobility will be significantly related to staff retention, a range of additional variables will also be analyzed for their impact on this outcome (e.g., age, socioeconomic status, work environment, shift schedules and workloads, morale and quality of care within the facility). A multiple regression analysis is utilized when several independent variables are analyzed to determine their impact on one dependent variable (Benjamin, Guttery & Sirmans, 2004). A predictive value can then be given to each independent variable based on the strength of its influence on this predetermined research outcome (Benjamin, Guttery & Sirmans, 2004). Additionally, a multiple regression analysis allows for researchers to determine the amount of variance each independent variable account for, based on the sample population used (Benjamin, Guttery & Sirmans, 2004).

In a standard multiple regression analysis, such as that used in the current study, all independent variables are entered into the regression model at the same time, and the model provides a coefficient value based on what each independent variable adds to the prediction of the main dependent variable (Ngo, 2012). For example, the current study hypothesizes that perceptions of upward mobility will have the highest predictive value for nursing staff retention of all the independent variables. Once the regression model has been developed based on the selection of key independent and dependent variables, the model can be tested for validity based on the implementation of a global F test, which determines the significance of all independent variables for predicting the dependent variable (StatSoft, n.d.). Assumptions must also be accounted for in the implementation of this data analytic procedure, including that the fact that there may be random error in the model, the error is normally distributed across the data set, and that random error exists independently within the data set (Ngo, 2012). Assuming no assumptions have been violated, the model is validated by examining the predictive values of each independent variable, exploring the parameters of the regression model to identify inconsistencies, and testing the model with alternative data sets (Ngo, 2012).

Rationale for Analysis Chosen

As described above, a multiple regression analysis was selected because of its appropriateness for the research question at hand. As the current study is designed to determine the predictive value of multiple independent variables on one dependent variable, a multiple regression analysis allows for the implementation of just one statistical test, rather than multiple correlations or t-tests comparing the differences between each independent variable (Benjamin, Guttery & Sirmans, 2004). This method of analysis is a relatively simple and accurate way of determining the degree to which multiple factors predict one specific outcome (Burrill, n.d.).). Additionally, such analysis can be used to determine the predictive value of other variables within the model when one or more independent variable is removed (Burrill, n.d.). For example, this test might be used to determine the impact of work environment on staff retention when perceptions of mobility are no longer a factor. As the aim of the current study is ultimately to conceive strategies for improving nursing staff retention, a multiple regression analysis will provide valuable insight into the various contributing factors that impact this outcome (Burrill, n.d.).

An additional factor used in the selection of multiple regression analysis for the current study is that many similar studies have relied on the same analysis. For example, Taunton, Boyle, Woods, Hansen and Bott (1997) conducted a classic study exploring the impact of manager leadership in the retention of hospital staff nurses. In this study, multiple factors related to manager leadership were used to determine their ability to predict staff retention. Multiple regression analyses revealed that predetermined theoretical variables accounted for about 22% of the variance in retention rates (Taunton et al., 1997).

In a more recent study, Laschinger, Leiter and Gilin (2009) used a multiple regression analysis to determine the impact of workplace empowerment, incivility, and burnout on nursing staff retention. Results from this analysis determined that these variables were associated with relatively high predictive values, yielding several promising directions for future researchers and practitioners to explore (Laschinger, Leiter & Gilin, 2009).

Interpretation of Results to Assist in Answering Research Question

When interpreting results from the statistical procedures described above, a range of implications can be made that assist in answer the research question at hand. Specifically, the regression model will provide a correlation coefficient for each independent variable entered into the model (Benjamin, Guttery & Sirmans, 2004). As the study hypothesis is that higher perceptions of upward mobility will be the most predictive variable in determining nursing staff retention, results from this statistical test will attribute a specific predictive value for this variable.

For example, if the test determines that perceptions of upward mobility are associated with a correlation coefficient of 0.7, one would be able to infer that 70% of all nursing staff retention is accounted for by the belief nurses hold about their chances for promotion. This represents a very high predictive value and would suggest that hospitals and healthcare researchers must place a much greater emphasis on increasing these perceptions in order to retain nursing staff (California State University Long Beach n.d.). On the other hand, if the model demonstrates that the predictive value for perceptions of upward mobility are relatively low (e.g., 0.3 or less), these professionals may more wisely spend their time emphasizing other work-related factors, such as the work environment or the quality of care provided by the facility in which the nurse is employed (California State University Long Beach, n.d.). As can be seen, a multiple regression analysis is a highly versatile test that yields a plethora of information with minimal effort (Burrill, n.d.). While the model is correlational in nature and does not provide causal evidence as to the relationship between variables, it accurately compares numerous potential contributing factors with each other and highlights the most important variables for predicting specific research outcomes (Benjamin, Guttery & Sirmans, 2004). Multiple regression analysis can also be used as a preliminary step in more sophisticated statistical procedures, such as factor analysis. Numerous researchers, however, limit their analytic procedures to just multiple regression analysis because of its simplicity and interpretive ease (Burrill, n.d.).

References

Benjamin, J. D., Guttery, R. S. & Sirmans, C. F. (2004). Mass appraisal: an introduction to multiple regression analysis for real estate valuation. Retrieved from: http://www.real-analytics.com/finc_674/mass%20appraisal.pdf. Accessed 23 September 2013.

Burrill, D. F. (n.d.). Modeling and interpreting interactions in multiple regression. Retrieved from: http://www.minitab.com/uploadedFiles/Shared_Resources/Documents/Articles/interactions_in_multiple_regression.pdf. Accessed 23 September 2013.

Cangelosi, J., Markham, S., Bounds, W. (1998). Factors related to nurse retention and turnover. Health Marketing Quarterly,15(3), 25-43. Retrieved from  http://ncbi.nlm.gov/pubmed/10181445

Carayon, P. & Gurses, A. (n.d). Nursing workload and patient safety—a human factors engineering perspective. Retrieved from: http://www.ahrq.gov/professionals/clinicians-providers/resources/nursing/resources/nurseshdbk/CarayonP_NWPS.pdf.

California State University Long Beach (n.d.). Multiple regression. Retrieved from: http://www.csulb.edu/~msaintg/ppa696/696regmx.htm. Accessed 23 September 2013.

Dunn, R. & Haimann, T. (2010). Healthcare Management. Chicago: Health Administration Press.

Haase, R., Schultheiss, T., Kempocke, R., Thomas, K. & Ziemssen, T. (2012). Use and acceptance of electronic communication by patients with multiple sclerosis: a multicenter questionnaire study. Journal of Medical Internet Research, 14(5), 135-151.

Hsieh, F. Y., Bloch, D. A. & Larsen, M. D. (1998). A simple method of sample size calculation for linear and logistic regression. Statistics in Medicine, 17, 1623-1634.

IBM (2013). IBM SPSS statistics editions. Retrieved from: http://public.dhe.ibm.com/common/ssi/ecm/en/yts03009usen/YTS03009USEN.PDF. Accessed 23 September 2013.

Ivers, N. M., Taljaard, M., Dixon, S., Bennett, C., McRae, A., Taleban, J., ... (2011). Impact of CONSORT extension for cluster randomised trials on quality of reporting and study methodology: review of random sample of 300 trials, 200-8. BMJ, 343. Retrieved from: http://www.ncbi.nlm.nih.gov/pmc/articles

Laschinger, H. K. S., Leiter, M. & Gilin, D. (2009). workplace empowerment, incivility, and burnout: impact on staff nurse recruitment and retention outcomes. Journal of Nursing Management, 17(3), 302-311. Retrieved from: http://onlinelibrary.wiley.com

Lobiondo, G. & Haber, J. (2005). Nursing Research: Methods and Critical Appraisal for Evidence-Based Practice. St Louis, MO: Mosby-Elsevier.

Lyberg, L., Biemer, P., Collins, M., Leeuw, E. D., Dippo, C., Schwarz, N., ... (Eds.) (2012). Survey Measurement and Process Quality (3rd ed.). New york: John Wiley & Sons, Inc.

McCann, T. & Clarke (2003). Grounded theory in nursing research: part 1-methodology. Nursing Research, 11(2), 7-18.

Ngo, T. H. D. (2012). The steps to follow in a multiple regression analysis. Retrieved from: http://support.sas.com/resources/papers/proceedings12/333-2012.pdf. Accessed 23 September 2013.

Platt, R., Takvorian, S., Septimus, E., Hickok, J., Moody, J., Perlin, J., ... (2010). Cluster randomized trials in comparative effectiveness research: randomizing hospitals to test methods for prevention of healthcare-associated infections. Medical Care, 48(6), 52-57.

Polit, D. & Beck, C. (2010). Generalization in quantitative and qualitative research: myths and strategies. International Journal of Nursing Studies, 47(11), 1451-1458.

Portaluppi, F., Smolensky, M. H. & Touitou, Y. (2010). Ethics and methods for biological rhythm research on animals and human beings. Chronobiology International, 27(9-10), 1911-1929.

Shader, K., Broome, M., Broome, C., West, M. & Nash, M. (2001) Factors influencing satisfaction and anticipated turnover for nurses in an academic medical center. Journal of Nursing Administration, 31(4), 210-216.

Shields, M. & Ward, M. (2001). Improving nurse retention in the National Health Service in England: the impact of job satisfaction on intentions to quit. Journal of Health Economics, 20(5), 677-701.

StatSoft (n.d.). How to find relationship between variables, multiple regression. Retrieved from: http://www.statsoft.com/textbook/multiple-regression/. Accessed 23 September 2013.

Stern, P. (1980). Grounded theory methodology: its uses and processes. Image, 12(1), 20-23.

Taunton, R. L., Boyle, D. K., Woods, C. Q., Hansen, H. E. & Bott, M. J. (1997). Manager leadership and retention of hospital staff nurses. Western Journal of Nursing Research, 19(2), 205-226. Retrieved from: http://wjn.sagepub.com/content/19/2/205.short. Accessed 23 September 2013.

Touvier, M., Mejean, C., Kesse-Guyot, E., Pollet, C., Malon, A., Castebon, K., ... (2010). Comparison between web-based and paper versions of a self-administered anthropometric questionnaire. European Journal of Epidemiology, 25, 287-296.

Turvey, C., Sheeran, T., Dindo, L., Wakefield, B. & Klein, D. (2012). Validity of the Patient Health Questionnaire, PHQ-9, administered through interactive-voice-response technology. Journal of Telemedicine and Telcare, 18(6), 348-351.

U.S. Department of Health and Human Services (n.d.). Guidelines for responsible data management in scientific research. Retrieved from: http://ori.hhs.gov/images/ddblock/data.pdf. Accessed 16 September 2013.

West, M., Guthrie, J., Dawson, J., Borrill, C. & Carter, M. (2006). Reducing patient mortality in hospitals: the role of human resource management. Journal of Organizational Behavior, 27(7), 983-1002.