find a recent (no older than 3 years) article reporting the results of a nursing or health research study in which REGRESSION analysis is used (try using health research and correlation as search terms).
APPRAISAL OF REGRESSION ANALYSIS OF A RECENT ARTICLE ON NURSING
Name of the Student
Appraisal of Regression Analysis of a Recent Article on Nursing
Study Selected for Appraisal
Yang., L, Dan., C, Xuemei., Z, Qiul, Z, Ningning., X & Xiaoying, S.(2014). Perspectives from Nurse Managers on Informatics Competencies. The Scientific World Journal. Vol. 2014. http://dx.doi.org/10.1155/2014/391714 Brief of the Study
Nurse Managers play an important role in providing care to patients with different health care needs. Hence, accessing appropriate information is one of the pre-requisites of providing effective care for patients. With an increase in electronic health record documentation and increased use of information technology in nursing practice, nurse managers must develop adequate competency in accessing such informatics.
The present study was done to evaluate the factors that are responsible for increasing the information competency in nurse managers. The authors used multiple regression equation to predict the competency of nurses in using various informatics systems in healthcare. The independent variables considered in this study were a level of education, experience and training on informatics. The dependant variable in this study was information competency. The independent variables were considered as “dummy variables” and a numerical magnitude was assigned to the qualitative aspects of the independent variable.
Methodology Adopted & Regression Model Selected
Multiple regression equation refers to a predictive equation that helps to derive the magnitude and direction of a dependant variable from the magnitudes of independent variables considered together. Hence, regression equation produces the likely values of a dependant variable from the likely values of single or multiple independent variables. The co-efficient of each independent variable signifies the correlation co-efficient of that independent variable with the dependant variable. When the correlation co-efficient is positive, it indicates increasing the magnitude of the independent variable will increase the magnitude of dependant variable.
On the other hand, if the correlation co-efficient is negative; it indicates that decreasing the magnitude of the independent variable will increase the magnitude of dependant variable. The p-value of the correlation co-efficient or the regression equation as a whole indicates the significance of correlation and significance of the overall regression equation (Robinson, 2001).
Assumptions and Statistical Interpretation
The null hypothesis (H0) contends (if p >0.05) that there is no significant correlation between independent variables and dependant variables and any correlation or equation of relation has happened due to chance factors of random sampling or from bias (Polit & Beck, 2004). This indicates in 100 observations, more than five observations has happened due to chance and the so null hypothesis is accepted.
On the other hand alternative hypothesis or experimental hypothesis contends (if p < 0.05) that there is significant correlation between independent variables and dependant variables and any correlation or equation of relation has not happened due to chance factors of random sampling or from bias. This indicates in 100 observations, only less than five observations have happened due to chance that is considered too low. Hence null hypothesis is rejected, and the experimental hypothesis is accepted (Plichta & Kelvin, 2013).
The results indicated that the level of education and training on informatics were positively correlated to Informatics competency. The correlation co-efficient of both the independent variables with dependent variable was 0.535 and 0.234 respectively with p-values for both were less than 0.05. On the other hand experience of nursing was negatively correlated, as the correlation co-efficient was -0.62. This observation was also statistically significant (p<0.05).
The co-efficient of determination (R2) was 72.8%. It means that information competency is chiefly determined by these three independent variables considered in the study. However, there could be other possible independent variables that may influence information competency in nurse managers.
The study indicated that level of education and training on informatics improves informatics competency in nurse managers. However, the experience of nursing shows to be negatively correlated which may be due to increased age of nurse managers’ leads to decreased level of information competency.
ReferencesPlichta,S., & Kelvin, E. (2013). Munro’s statistical methods for health care research
(6th ed.). Philadelphia, PA: Lippincott Williams & Wilkins.Polit, D. F. & Beck, C.T. (2004). Nursing research: principles and methods. (7th ed.).
Philadelphia, PA: Lippincott, Williams & Wilkins.Robinson, J.H. (2001). Mastering research critique and statistical interpretation – guidelines
and golden rules. Nurse Educator, 26, 136-141.