Patient Satisfaction and Nursing

¶ … size is an important step in the sample selection process. In Negarandeh, Bahabadi & Mamaghani's (2014) study, the procedure used to determine the sample size is clearly described. Following a pilot study, using the sample size formula, and based on the population of the hospital in which the trial was carried out, the authors estimated the sample size at 100 participants (50 participants in each group). Revealing how the sample size was determined enables readers to ascertain the extent to which the sample selected is representative of the larger population (Thomas, 2009). In quantitative research, a representative sample is important for improving the generalizability of findings (Bryman, 2008). In this case, the sample was quite representative of the study population. The study setting was a 530-bed hospital. The trial was specifically conducted in the medical surgical ward, which had 40 active beds, meaning the findings can readily be generalized to the study population. The next step in the sampling process entails recruiting participants, which may involve random or non-random sampling techniques. In this case, subjects were recruited through convenience sampling. As convenience sampling is a form of non-random sampling, some bias may have occurred (Kothari, 2004).

Negarandeh, Bahabadi & Mamaghani's (2014) article provided valuable insights about sample selection. This knowledge will be important for my future DNP scholarly projects. For me to generate credible findings in my future DNP projects, I must know how to determine the right sample size and the appropriate technique to use in locating the sample. Also, by reading the article, I was able to connect my previous knowledge with a real-world example. This increased my understanding of sample size determination and selection. Furthermore, the article demonstrated the importance of documenting sample size determination and sample selection procedures. In my future DNP projects, I must pay attention to disclosing these elements.

Part II

The aim of Negarandeh, Bahabadi & Mamaghani's (2014) study was to examine the impact of nursing rounds on patient satisfaction with nursing care. The study specifically focused on two variables: patient satisfaction (the dependent variable) and regular nursing rounds (the independent variable). Patient satisfaction was measured using the Patient Satisfaction with Nursing Care Quality Questionnaire (PSNCQQ). A number of studies have confirmed the validity and reliability of the tool (e.g. Ksykiewicz-Dorota et al., 2011; Al-Abri & Al-Balushi, 2014). In Ksykiewicz-Dorota et al.'s (2011) study, for instance, the internal consistency of the tool was 0.96 on Cronbach's alpha, with predictive validity ranging from 0.57-0.84. This shows that the tool was appropriate for measuring the variables of interest. The data collection process involved administering the PSNCQQ tool on the second and fifth days of hospitalisation. To ensure effective data collection, the nurses involved were first taken through a twenty-minute training session. The length of the training session may seem short, but I believe it was adequate given the nature of the study.

One thing that stood out for me is the training of data collectors prior to embarking on the process of data collection. For me, this was new knowledge. Often, clinical researchers rely on hospital personnel to collect data. As demonstrated in Negarandeh, Bahabadi & Mamaghani's (2014) study, training data collectors is crucial for ensuring the required data is collected as desired. This will be a major point of consideration in my future DNP projects. Training data collectors, however, does not necessarily mean that the researcher leaves everything to them. The researcher must closely supervise the data collectors to ensure they comply with the provided guidelines throughout the study period. Nevertheless, too much researcher supervision may result in a Hawthorne effect (Negarandeh, Bahabadi & Mamaghani, 2014). The data collection part of the article was also an opportunity for me to connect previous knowledge with a real-world study.

Part III

It is not clear in Negarandeh, Bahabadi & Mamaghani's (2014) study whether data cleaning and outlier detection was done as the authors did not mention these aspects. In research, data cleansing is important for eliminating inaccurate, incomplete, invalid and irrelevant data (Broeck et al., 2005). For instance, some subjects may leave some questionnaire items unanswered, or may answer some items wrongly. Eliminating such data ensures the researcher reports more accurate and consistent findings (Denscombe, 2010). It is also essential to check outliers during data analysis. This enables the researcher to check for values that lie outside the normal distribution. The authors also did not mention if there were any missing data and how they handled it. Often caused by factors such as non-response, missing data can be handled through imputation (substituting the missing data with values), interpolation (constructing new data), and deletion (reducing the data) (Bryman, 2008). An attractive aspect of Negarandeh, Bahabadi & Mamaghani's (2014) study is that there was no attrition at all -- data from all the 100 participants recruited at the beginning was included in the final analysis. 0% attrition is quite unusual in clinical trials (Maltby et al., 2015).

Though the response rate was noticeably outstanding, the data analysis section of the article has significant limitations. For me, these limitations particularly stood out. As mentioned earlier, the researchers did not clarify whether there were instances of incomplete or missing data. Even though there may be no attrition, instances of incomplete responses may often be encountered (Robson, 2016). Including this information would have placed the reader in a better position to assess the quality of the findings. In my future DNP projects, it would be vital to avoid this limitation by disclosing as many details about the data analysis process as possible. On the whole, compared to other sections, the data analysis section in Negarandeh, Bahabadi & Mamaghani's (2014) article did not really provide any new learning for me.

Part IV

In Negarandeh, Bahabadi & Mamaghani's (2014) study, descriptive statistics were used to report the characteristics of the sample. More specifically, frequencies were used to report demographic attributes, especially with respect to age, gender, marital status, educational status, and history of hospitalisation. Means were used to report patient satisfaction scores. Inferential statistics were also used. As mentioned before, the study aimed to examine the impact of regular nursing rounds on patient satisfaction. Chi-square test, Fishers exact test, and independent t-test were performed to examine the relationship between the two variables. Given the nature of the study (Denscombe, 2010), the inferential statistics used were appropriate. P values of less than 0.05 were reported, meaning that the results were statistically significant. In other words, regular nursing rounds significantly increase patient satisfaction in nursing care.