Statistical Articles Analysis - Research Paper Example

Analysis of Research Articles al Affiliation Analysis of Research Articles Introduction Conceptually, this papers analysestwo statistical research articles by summarizing all the statistical parameters such as the assumptions, research questions, and the hypotheses. In addition, the paper discusses on the methodologies used, particularly, the sampling method, sample size, variables used, and data collection, from which it evaluates how applicable and relevant the authors of the articles used the techniques (Goodman, 2011). Further, the paper evaluates how the authors analyzed their data from where the discussions were based and conclusions drawn. Explicitly, in order to appositely rigor on the statistical mechanisms in the articles, it is prudent to outline the study limitations that could have resulted to deviance from the recommended techniques. In sum, this analytical paper gauges how all elements of research articles are employed in the literatures the make apt recommendations or suggestions where necessary. Hence, the following two articles form the basis of analysis or evaluation of this paper: 1. Addison, K., & Luparell, S. (2014). Rural Nurses Perception of Disruptive Behaviors and Clinical Outcomes: A Pilot Study. Online Journal of Rural Nursing and Health Care, 4(1), 66-82. Statistical Summary In the article by Addison and Luparell it reports on research conducted in two rural hospital facilities in order to relate two major parameters; that is, disruptive behavior of the staff and the patient outcomes. Explicitly, disruptive behavior of nurses and physician represents the independent variable that is tested against many other dependent variables such as patient safety, mortality rates and so on (Addison, K., & Luparell, 2014). However, from the article, it is evident that the authors held several assumptions that are implicitly depicted within the report. Methodologically, the researchers employed sampling techniques to define their study area and the sample size. Through surveying, electronic questionnaires sent via the email of the hospital staff resulted to relatively high response rate despite the various limitations that barred apposite data collection. Markedly, the study area was a rural setting; however, it is clearly that the area sampled was not well representative of the whole region where the results need to be extrapolated. Notably, the article depicts how the authors used relational regression model to analyze their data; for example, the independent variable, disruptive behavior, was regressed against several dependent variables such as patient outcomes, patient safety, communication in hospitals, and staff concentration. Such regression heralded results that allowed apparent correlation of the variables and to help the reader visualize the relations between the nurses and physicians. In addition, factor analysis enriched the relational study and to permit comparison and contrast of the study results with those of the prior research conducted in an urban setting. Explicitly, the researcher presented their resulted clear and well labeled tables and bar graphs so as to enhance visualization and conveyance of the information (Addison, K., & Luparell, 2014). Undoubtedly, the article had well-grounded background study and purpose, and ultimately the authors provided the answer to the research questions despite the present of several limitations and hitches to the study. Evaluation of the Statistical Elements in the Article Assumptions of the Study As aforementioned, the article implicitly holds various assumptions. For example, the researchers assumed that all nurses and physicians in the rural Montana had access to personal computers through which the questionnaires were sent. Unfortunately, it became clear during and at the end of the study that most of the hospital staff lacked the computers. Addison and Luparell explains this by reconsidering their assumption as the unexpected limitation of the research. Again, during the survey, it was assumed that the response is only influenced by personal voluntarism; however, that factor impacted on the level of participation. It was implicit that a number of potential participants were barred due to fear of disclosure of the information to their seniors since the survey diverted to using shared hospital computers. Additionally, there was an assumption in the definition of a rural hospital facility and territory so as to define the scope of the study. Demographically, it was assumed that the population of the rural territory should be less than 50,000 people; however, it was not clear on whether the figure is the population of hospital nurses and physicians or the general population. Hence, it is evident that some of the assumptions were not properly classified (Addison, K., & Luparell, 2014). Also, the study assumed that rural facilities had a capacity of not more than 125 beds, and this is correctly tested during the sampling process of two hospitals. Review of the Data Analysis Notably, the data from this study were analyzed both qualitatively and quantitatively. In essence, qualitative concepts were used to carry out factor analysis while quantitative values were used to enhance the relational regression and enrich the substantial qualification of data. An analysis was done to test the hypotheses or to answer the several research questions and to test the relationship between the variables. Data analyzed was collected through the use of questionnaires during the survey that employed cross-sectional sampling techniques. In particular, the aforementioned aspects are discussed below(Goodman, 2011). The Research Questions Appropriately, the study had three-research questions that limit the complexity of research caused by wider scope. The first question asked about the incidences of disruptive behaviors in nurses found in the rural hospitals of Montana. Explicitly, the question requires an answer that is tested in terms of frequency; however, it poses a challenge in framing a specific question in the questionnaire. Second question asked on whether there is a connection between disruptive behavior and patient safety, adverse effects, and medical errors. Therefore, the second question calls for testing the correlation or causation factor between the variables. Lastly, the third question is a form of a null hypothesis, where the researchers ask if nurses perceive a negative relationship between the independent variable and behavioral and psychological variables like team collaboration, communication, frustration, and stress. Evidently, the data analyzed answered all the questions, but the only limitation how to differentiate causation from correlation (Addison, K., & Luparell, 2014). The Methodology (Data Collection and Sample Size) Collection of data was conducted by the use of questionnaires similar to those used in a prior study in VHA West Coast hospital. The questionnaires had 21 questions, which included yes-no answers, multiple choices, open-ended questions, 5, and 10-point scale queries. They were remitted via the emails of the nurses; reminder sent on the 15th-day, and they remained online for up to 30 days. Obviously, a preliminary consent letter was sent to seek participation from the potential respondents. Technically, a descriptive design of cross-sectional sampling was used. First permission was sought from the relevant bodies and five rural hospitals contacted for participation, out of which, only two confirmed their participation. One of the hospitals was denoted as a CAF-critical access facility reporting 11-20 beds. The second hospital was small reporting 51-100beds. Eventually, 120 nurses from the two hospitals got the invitation, where 57 surveys resulted to a response rate of 47.5% out of which 29% from CAF, 70.9% from 51-100 beds hospital, and 1% with no identified working setting. Critically, electronic deterrents, non-consents from three hospitals, and the consequential small sample size limit the relevance of the results (Addison, K., & Luparell, 2014). Since the sample size is proportional to calculation of power, a small sample size did not allow proper power analysis. However, in order to calculate the probability of adopting one of the two hypotheses-null or alternative, the author did some statistical test calculations. Surely the assumptions of the study were met by changing the research perception into proven facts; however the measurement was not that appropriate. It was because the relatively non-representative samples size lead to omissions or multiple assumptions that corrupted the reliability of the results (Goodman, 2011). Variables Independent and dependent variables are revealed implicitly and explicitly. For example: Independent Variable · Disruptive behavior Disruptive behavior of nurses was tested against the multiple dependent variables. Qualitatively, the independent variable was defined as the actions, activities, or utterances of the nurses that had effects on patient outcomes and relationship among the co-workers Dependent variables Disruptive behavior was used to test several dependent variables such as: · Patient outcomes · Adverse events · Patient safety · Team performance · Work concentration · Communication Data Display In order to facilitate proper visualization and comprehensive presentation of the results, the researchers used clear and well labeled tables and bar graphs. For example, to display the percentage of nurses and physicians who showed disruptive behavior, the percentage of respondents was plotted against the cases of disruptive behaviors in nurses and physicians in a bar graph. Ideally, the graph reveals the relational frequencies of disruptive behavior, where it was evident that nurses had higher percentages as compared to the physicians. Again, another bar graph was drawn to present YES or NO answers to some selected questions to show the percentages of the respective responses. Furthermore, a matrix table was used to record the results in correspondence to the variables. The table answers the question on how often does the independent variable result to behavioral and psychological effects such as stress, reduced relations, loss of concentration, frustration, reduce communication and so forth. Responses are substantiated as never, rarely, sometimes, and so forth, from which quantitative percentages were matched in the cells. Similar tabulation was used to relate disruptive behavior to other outcomes. Undoubtedly, the tables and graphs comprehensively present the relationships between the variable in the most visual and understandable notions. Evaluation of the Data Analysis The acclaimed author of the article conducted commendable data analysis; however, some aspect of the data they analyzed would limit interpolation and extrapolation of the results to the whole population. Contextually, the analysis revealed that nurses in rural setting experienced or showed high-disruptive behaviors than the physicians, and cumulatively more than their urban counterparts. In addition, it was evident that the independent variable leads to a chain or network of outcomes, which limits distinctiveness and understanding the causative variable (intermediary variable). In their conclusion, the results of the study were consistent with those of prior studies conducted in an urban context; however, the percentages of disruptive behaviors are higher in rural hospitals. Again, the disruptions are frequent in nurses than physicians and that disruptive behavior results to adverse events, impaired staff relationships, and decreased patient safety. In reality, the first conclusion is not right since urban hospitals have been proven to show much disruption. Nonetheless, the second and third conclusions are consistent with the real scenarios. For example, nurses do manifold tasks in hospitals, which correlate to increased disruptive behaviors(Chiswick, 2002). Overall, the authors have done well in giving the grounds and the rationale of the study, description, measuring, and testing of the variables. However, the methodology, particularly the estimation of the sample size is poorly done. Hence the trust on the results is high on specific setting, the rural, but non- representative of the whole population. Limitations of the Study · Face and contrast validity were not aptly done on the research questions · Most of the nurses lacked personal computers or infrequently check their emails · Unrepresentative sample size that limit interpolation or extrapolation and power analysis · Potential of cofounding variables due to multiple relationships of variables · Survey questions did not properly define disruptive behavior Understanding the Data Conclusively, this study is very critical for nursing students, since it gives the limelight on how disruptive behaviors adversely affect the relationship of co-working nurses and the patients’ outcomes; the primary objective of nursing. Peer students might feel that the study has a social perspective, which may be a confusing thing; however, they should understand that nursing involves social, emotional, psychological, and medical support and advocacy (Chiswick, 2002). Thus, this study is integral. Nevertheless, the students could also question the first conclusion that compares the disruptive behavior in rural and urban setting. In summary, the research provided sufficient evidences to shape nursing career. For example, the use of one independent variable that influences multiple effects should serve as a wake-up call. All told, only one issue that affects nursing results to manifold adverse outcomes. 2. Kroger-Jarvis, M. (2014). Evaluating Prostate Cancer Knowledge in Rural Southeastern Indiana County. Online Journal of Rural Nursing and Health Care, 14(1), 83-99. Statistical Summary Kroger-Jarvis (2014) reports on a qualitative study purported to carry out a descriptive analysis so as to demonstrate the relationships between knowledge and prostate cancer screening. Other variables to be related to the screening included spouse and health care provider’s influence. Contextually, the study was conducted in men of Ripley County in Southeastern Indiana. Preliminarily, findings indicated a relationship between prostate screening, men’s knowledge, marital status, and health provider’s influence. Notably, the objectives of the research are answered by relating the variable and the author also does well in setting and confirming the assumptions. Despite some statistical limitations, the study makes applicable conclusions. Assumptions The main assumption was that men seek screening depending on prior information. Markedly, the study goes ahead to back the assumption by using descriptive analysis to assert a relationship between prior knowledge and to go for screening. In addition, the author explains that the knowledge is enhanced by marriage, recommendation from medical practitioners, and family history on prostate cancer. Again, the study assumes that rural men are less knowledgeable than their urban counterpart, which is also proven right because urban settings are associated with interactions, better hospitals, and quicker information access and exchange. Review of the Data Analysis Descriptive analysis of the qualitative and quantitative data collected from systematic literature review -secondary data and primary investigation. The primary investigator (PI) developed 11-item surveys, then distributed in retail sites, eateries, and two family practices. Through sampling, the study made the comparison of the demographic data by using descriptive statistics together with parametric evaluation that used Chi-square Association Test. In order to enable reliability, face validity test was done on the research questions (Lafuente-Millán, 2012). The Research Objectives Purposefully, the research was to find answers to four questions; first, to evaluate if men feel that they have information on screening, which is proved during the survey. Second, was evaluating men’s knowledge; methodologically, the research does this by asking them on the screening guidelines and if they know the recent changes concerning the disease. Third, was to evaluate where rural men obtain the information about cancer screening. Evidently, medical practitioners, spouses, and family history on prostate cancer were the main determinants of the men’s knowledge. Lastly, was to evaluate if already men have obtained the knowledge or if they are planning to. Notably, (37%, n=59) of the mens participants had the knowledge concerning screening while about 55% were planning for doctor’s appointment (Kroger-Jarvis, 2014). The Methodology (Data Collection and Sample Size) Descriptively, the research exhibits a needs assessment in a survey methodology in order to evaluate the relationship that link men’s knowledge and prostate cancer screening in the rural setting of Ripley County in Southeastern Indiana. A survey consisting of 11 items was developed then distributed to retail sites, eateries, and in two family practices. The 11 items were prostate cancer guidelines used to measure the knowledge of men about the disease. In order to enhance the validity, face validity was measured in the research questions by asking the study’s committee members whether the questions are useful or well constructed. Again, power analysis is done through non-parametric evaluation of the statistical probabilities of the outcomes using a Chi-square test. Sampling of the study had apt criteria of inclusion and exclusion during the recruitment of the research participants (Wegman, 2012). Inclusion of men involved those 50 or above years of age, must reside in the rural Ripley County, able to speak, write, and read in English, and consented to the information sheet and the flyer in order to fill the research questionnaire. Conversely, exclusion was for men who: were non-English speakers, cognitively impaired, illiterate, and under the age of 50 years. Statistically, the rural Ripley is a non-metropolitan area with a population of about 25,583, the study area from which the sampling frame was derived (Kroger-Jarvis, 2014). As aforementioned, recruitment of respondents was done by principle investigator (PI), posters, and flyers. Obviously, prior permission had to be sort for; for example, written permission to doctors’ offices while verbal permissions from retail establishments and eateries. If permission was granted, the posters were stuck on physician doors and walls and secured on tables in case of retails and eateries. In addition, flyers and posters, which provided the research information and the contact information, were sent via the telephone or emails. Importantly, the inclusion and exclusion criteria also explained the risks and benefits of participating in the study. Participants were reminded on voluntary participation, and they could choose to refuse, withdraw, or participate at any time, and all the information given was anonymous. Collection of data was done by questionnaires with 11 items, which provided the guidelines for measuring the men’s knowledge on prostate cancer screening. The items were based on the American cancer Society (ACS) new guidelines on prostate cancer. Objectively, the guidelines asked measured men’s knowledge, where they obtained it, who or what influenced them, and if they are planning to go for cancer screening. Conceptually, having the information of the risks, benefits, and where to go for cancer screening is linked to men making informed decisions when going for a screening. Variables Independent Variables · Men’s’ knowledge on prostate cancer screening · Demographic data such as marital status, age, internet access, education levels, employment, and residence. Dependent Variables · Making informed decision (IDM) to go for prostate cancer screening · Knowledge on prostate cancer screening Note: it is paradoxical that men’s knowledge appears as both dependent and independent variable, for example, positive demographic information influence men’s knowledge, as the knowledge influence going for cancer screening. Data Display Nicely, the author displays the demographic characteristic of the participant in a table in order to allow proper visualization and presentation. Again, most of the main responses to the questions are tabulated, where the research institutes the quantitative and qualitative statistics. Evaluation of the Data Analysis The researcher does well in measurement of the variable, questions, and objectives of the study. For example, conducting face validity of the study questions. Implicitly, the study used demographic parameters to answer the objective, which led to proper relational study of the variables and testing for the hypothesis. Additionally, the relational regression of the variables is enhanced by descriptive analysis that employed non-parametric evaluation that entailed the use of Chi-square Association test (Lafuente-Millán, 2012). Unfortunately, the study area is confined to rural Ripley, which limits the extrapolation. Results show that, (n=59), 37% men had knowledge, 54% had doctors’ appointments, 36% already screened, 55% plan, 83% recommended men of 50 and above to go for screening, but 66% said no need. In reality, being a rural area the research is consistent (Kroger-Jarvis, 2014). Markedly, the study did well in answering the objectives, inclusion recruitment, and measurement. For example, through non-parametric testing statistical probabilities (p) are measured like it was p=0.36 for the relationship between men’s knowledge and screening (Lafuente-Millán, 2012). However, exclusion criteria of participants are poorly done. Limitations of the Study · No construct reliability of validity on the survey · Associative relationships might lead to cofounding variables · The study is only confined to rural Ripley County and poor recruitment limit applicability of the results · Knowledge level is not measured in the survey questions Understanding the Data In conclusion, the study is crucial for nursing students in understanding how their recommendations influence patients’ going for cancer screening, particularly in the rural area (Chiswick, 2002). However, students might ask the questions concerning the research limitations, but perfection is like infinity. In sum, the researcher used proper measurement techniques and data analysis, which build high trust on the reliability of the results. References Addison, K., & Luparell, S. (2014). Rural Nurses Perception of Disruptive Behaviors and Clinical Outcomes: A Pilot Study. Online Journal of Rural Nursing and Health Care, 4(1), 66-82. Chiswick, M. (2002). Writing A Research Paper. Current Paediatrics, 12(5), 414-418. Duvvuru, A., Radhakrishnan, S., More, D., Kamarthi, S., & Sultornsanee, S. (2013). Analyzing Structural & Temporal Characteristics of Keyword System in Academic Research Articles. Procedia Computer Science, 20, 439-445.Chiswick, M. (2002). Writing A Research Paper. Current Paediatrics, 12(5), 414-418. Goodman, A. (2011). Emerging topics and challenges for statistical analysis and data mining. Statistical Analysis and Data Mining, 4(1), 3-8. Kroger-Jarvis, M. (2014). Evaluating Prostate Cancer Knowledge in Rural Southeastern Indiana County. Online Journal of Rural Nursing and Health Care, 14(1), 83-99. Lafuente-Millán, E. (2012). A Contrastive Study of Generic Integrity in the Use of Attitudinal Evaluation in Research Articles Written for Different Audiences. Brno Studies in English, 38(2), 79-96. Wegman, E. J. (2012). Special issue of statistical analysis and data mining. Statistical Analysis and Data Mining, 5(3), 177-177. Read More