Secondary Data Analysis, Research Paper Example

Abstract

This work in writing examines the issues surrounding qualitative and quantitative nursing research and examines specific cases in relation to these issues specifically in the areas of ethics, rigor, data collection, and analysis and in findings arising from such research in the area of secondary analysis.

Introduction

Secondary data analysis is defined by Polit and Hungler (2001) as involving “the use of data gathered in a previous study to test new hypotheses or explore new relationships.” (Stommel and Willis, 2004) There are many ins and outs when using secondary analysis in research. (Barrett, 1994) There are issues, which arise from the conduction of secondary analysis on “large, national survey data sets.” (Kniepp and Yarandi, 2001) Secondary analysis is used in research across many fields of study including such as criminology and nursing. (Riedel, 2000 and Orsi, et al, 1999). Building capacity in nursing involves the aspect of nursing research and the contribution that are made toward the body of knowledge in nursing. (Jeffs, et al, 2006)

The work of Jacobson, Hamilton, and Galloway (1993) entitled “Obtaining and Evaluating Data Sets for Secondary Analysis in Nursing Research” states that secondary analysis of existing data offers many advantages to the nurse researcher. Data from large-scale studies are such that may be “reanalyzed through secondary analysis with a fresh perspective, thus enhancing the original study’s contribution to scientific knowledge.” (Jacobson, Hamilton and Galloway, 1993) It is additionally related that there is a special level of care that must be exercised by the secondary analyst due to such as the “lack of involvement in data collection procedures may decrease insight intothe original study’s limitations, vigilant skepticism should accompany all phases of the research process in secondary data analysis, just as it should in all other research.” (Jacobson, Hamilton and Galloway, 1993)

The work of Bibb (2007) entitled “Issues Associated with Secondary Analysis of Population Health Data” reports that the study and analysis of population health data “are crucial to the success of population health.” Bierman and Bubolz (2003) note limitations using secondary data analysis, which includes analysis of population health data and the difficulty experienced in locating such data and the existing “incongruity of primary and secondary research objectives, as well as quality of data. (Paraphrased) The resolution of data location difficulties can be overcome through data location by searching “organization, survey and data warehouse web sites and through the review of published research studies.” (Clark and Cossette, 2000; and Smaldone and Connor, 2003)Various forms of secondary analysis exist including the: (1) cumulative meta-analysis; (2) meta-analyses; (3) meta-data analysis; (4) Meta-method; (5) meta-study; and (6) meta-syntheses. (Gough, and Land, 2004)

The work of du Plessis and Human (2009) reflects on the level of “meaningful research” that is available to nurses when conducting collaborate secondary analysis. Heaton (1998) in the work entitled “Secondary Analysis of Qualitative Data” reports that secondary analysis “involves the utilization of existing data, collected for the purposes of a prior study, in order to pursue a research interest which is distinct from that of the original work.”

Description

A recent review of trauma among veterans and military personnel reports that they “experience higher rates of trauma exposure in comparison to the general population, with associated higher rates of post-traumatic stress disorder.” (Suris and Lind, 2008, p.250) The type of trauma most widely studied is reported as “trauma in female veterans” due to sexual assault including such as “childhood sexual assault, civilian adult sexual assault and sexual assault that occurs while serving in the military.” (Suris and Lind, 2008, p.251) MST is defined by the Department of Veterans Affairs as “sexual harassment that is threatening in character or physical assault of a sexual nature that occurred while the victim was in the military, regardless of geographic location of the trauma, gender of the victim or the relationship to the perpetrator.” (Suris and Lind, 2008) MST is reported as the term used “to describe physical assault, sexual assault, stalking or harassment that occurs during active duty.” (Suris and Lind, 2008, p.251) It was reported by the VA in 2002 “approximately 1.7 million women and 1% of men have experienced MST.” (Suris and Lind, 2008, p.251) The work of Suris and Lind (2008) report that there is a prevalence of military sexual trauma (MST) and associated mental and physical health consequences with risk factors for MST stated to include age, enlisted rank, negative home life, and previous assault history. In addition, MST has been “associated with increased screening rates of depression and alcohol abuse, in addition to significantly increased odds of meeting criteria of posttraumatic stress disorder.” (Suris and Lind, 2008) MST has further been associated with “reporting increased number of current physical symptoms, impaired health status, and more chronic health problems in veterans.” (Suris and Lind, 2008, p.250) Sexual assault against women in the military is reported as being “a widespread problem.” (Himmelfarb, Yaeger and Mintz, 2006, p.837) The present study reviews the issues of military sexual trauma and the qualitative and quantitative research methodologies used in various studies, the questions surrounding qualitative versus quantitative research and the evidence supporting each.

Advantages of Secondary Analysis

Secondary data analysis is described as “additional analysis of data already available in some form.” (Brosnan, Eriksen and Lin, 2002, p.368) The major advantage of secondary data analysis is stated to be that the data “already are collected. Because the data are readily available in a useful form, the major expense of data collection is saved. In addition, a secondary data analysis may serve to identify problems to be explored further in future studies.” (Brosnan, Eriksen and Lin, 2002, p.369) Pubic national data sets have inherent advantages: (1) they contain information about a wide variety of patients,  risk factors, health states, and interventions and because of this researchers are not forced into limitations of data from one population or area of study but have the capacity to examine many issues of health; (2) since the same information has been collected on a routine basis for many years it is possible to examine the trends and identify the differences; (2) an institutional Review Board may waive the necessity to gain approval which serves to save a great deal of time since the data sets are available for public use. (Brosnan, Eriksen and Lin, 2002, p.369)

Limitations of Secondary Analysis

Stated, as one potential limitation of the use of national surveys is that they contain stratified data collection methods that make the requirement of complex statistical analysis. Secondary analysis of research studies tends to prevent complications as the primary study design may have flaws or the data may be too old or in the case when the variable of interest in the present study was never collected. It is necessary to make careful examination of secondary data prior to it being used. It is necessary to answer specific questions including the following stated questions:

• What wasthe purpose of the original study, and howdoes it match with the purpose of the secondaryanalysis?
• How and under what conditionswere the data collected?
• What wasdone to check the accuracy of the data?
• What measurement instruments wereused, and how reliable and valid were they?
• What was the unit of analysis (i.e., individualor group)? Public-use national data setsaddress these questions in a consistent andstandard manner in the extensive documentationthat accompanies them. (Brosnan, Eriksen and Lin, 2002, p.369)

Fit with New Research Question

As previously stated, it is necessary to examine secondary data in regards to ‘fit’ and specifically how the data was derived and for what purpose in previous studies and ensuring that it is applicable and appropriate for use in the present study and the variables and issues examined. Questions that can be used to guide the determination of which variables in the data set are appropriate in a present study are reported to include those as follows:

• How were the concepts defined?
• What was the unit of analysis?
• When was the information collected? (Brosnan, Eriksen and Lin, 2002, p.369)

It is reported that a software program was developed by the NCHS “called the Statistical Export Tabulation System (SETS) to assist users in the efficient extraction of data files from the CDROM.” The SETS is reported as being readable in Windows and NT4 operating systems and contains variable word components that allow students to format the file.” (Brosnan, Eriksen and Lin, 2002, p.370)

Following the selection of the variables which are chosen by name the SETS generates subsets of data files which format variables and values which are defined for export. Next, the files are directly downloaded from the Centers for Disease Control and Prevention without use of the SETS. The database is imported to the SPSS and the students are able to collaborate with a statistician at the school to ensure that the appropriate statistical testing is used and the analysis can then be conducted. Students receive encouragement to document their research process in a special procedure manual used for evaluation and replication. In addition, students are provided with a “criterion sheet to evaluate their own studies and critique the work of their peers. The course grade is reported to be based on a written report documenting the findings of the study in addition to an audiovisual presentation of the primary components of the study. This report is stated to be considered “the first draft of a manuscript for publication.” (Brosnan, Eriksen and Lin, 2002) Stated limitations to the use of the data sets and issues that should be noted are that the national surveys “use stratified data collection methods and mayoversample segments of the population.” (Brosnan, Eriksen and Lin, 2002, p.370)

It is reported that stratification and clustering “results in a larger variance between groups than would occur in random sampling, and, as a consequence, appropriate analysis of the data must adjust for this design effect.” (Brosnan, Eriksen and Lin, 2002) It is reported that the ‘Research Applications’ course enables students in developing the required skills and in gaining the needed confidence to accomplish secondary analysis of national public data sets which are touted as “an instructive vehicle for teaching students about the research process that will serve them well into the future.” (Brosnan, Eriksen and Lin, 2002, p.371)

Quantitative versus Qualitative

While the view of qualitative research has been “refuted repeatedly in the past several decades, the emphasis on evidence-based research is against relegating qualitative work to second-class status.” (Ryan-Nicholls and Will, 2009, p.71) Issues that are reported to create additional complications in regards to qualitative inquiry’s merit as a science are inclusive of “the insufficient boundary clarification between qualitative and quantitative methods” as well as “the diversity of qualitative methods, the artistic features of qualitative research, and the propensity to evaluate qualitative inquiry against the conventional scientific standard of rigor.” (Ryan-Nicholls and Will, 2009, p.71)

Qualitative inquiry is reported as being both “distinct and separate” from quantitative inquiry and each research, methodology is committed to differentiated research styles, differentiated epistemologies and differentiated representations. Qualitative inquiry, according to Bogdan and Bilken (1982) is held to have emerged as a response to “the inadequacies of quantitative inquiry.” (Ryan-Nicholls and Will, 2009, p.71) Quantitative research has its focus on “causal relationships described in terms of observation statements, verifications and prediction”, while qualitative research, has as its focus, the “exploration of human behavior and the search for understanding through people’s actions.” (Ryan-Nicholls and Will, 2009, p.71)

Quantitative research can be differentiated from qualitative research which makes use of such as “life histories, historical narratives, ethnographic prose, first-person accounts, and biographical and autobiographical accounts” in that quantitative research makes use of such as mathematical models and statistical tables and graphs, and often reports the research findings in impersonal, third-person style.” (Ryan-Nicholls and Will, 2009, p.72) Sandelowski (1986) proposed a framework for the examination of the similarities and differences that exist between the qualitative and quantitative approaches through incorporation of four factors that were proposed in 1981 by Guba and Lincoln relating to tests of rigor (1) truth value; (2) applicability; (3) consistency; and (4) neutrality. (Ryan-Nicholls and Will, 2009, p.72) The framework is repotted as being inclusive of “alternative criteria for assessing the rigor of qualitative research: (1) credibility; (2) fittingness or transferability; (3) auditability or dependability; and (4) confirmability. (Guba, 1981, Lincoln and Guba, 1985 in: Ryan-Nicholls and Will, 2009, p.72)

In the present, qualitative researchers are reported as being “more likely to reveal how rigor was attained and maintained.” (Ryan-Nicholls and Will, 2009, p.73) However, it is noted in the work of Gearing (2004) that there is a tendency “to state something was done without explication of how it was done.” (Ryan-Nicholls and Will, 2009, p.73) The work of Morse (2007) provides a description of how generalizability in qualitative work is attained “through development of meaning, not prescriptive use of rules” and insists that “concepts must be well developed to be recognizable in other settings.” (Ryan-Nicholls and Will, 2009, p.73) It was the insistence of Farmer et al (2006) that “the richness of description in qualitative work of often missing from their own methodological descriptions.” (Ryan-Nicholls and Will, 2009, p.74)

Data Collection

The work of Cohn, Jia and Larson (2009) entitled “Evaluation of Statistical Approaches in Quantitative Nursing Research” reports a study that was published in the Clinical Nursing Research journal that states the purposes of the study being: (1) the development of a tool to assess statistical methods; and (2) use of the tool to evaluate recently published quantitative research.” (Cohn, Jia and Larson, 2009, p.223) The statistical approaches utilized in 152 studies published in the five top impact-factor nursing research journals from September 2005 to August 2007. Studies reported are stated to have been high in quality with 45% of those scoring 80% and 100% and 22% achieving a 100% score. It is reported that predictors of high scores were “interdisciplinary authorship, a statistician, coauthor and the number of aims.” (Cohn, Jia and Larson, 2009, p.223) It was concluded that “studies published in high-impact-factor nursing journals are statistically sound and provide a solid foundation for evidence-based practice.” (Cohn, Jia and Larson, 2009) It is reported that both researchers and practitioners alike are challenged in the present drive toward practice that is evidence-based. (Cohn, Jia and Larson, 2009, paraphrased) With the integration of research and practice, it has become critically important that quantitative study designs be of a quality that is particularly high. It is possible to be misled by the study conclusions in the case of incorrect statistical analysis or misinterpretation of that analysis. (Katz, Crawford, Lu, Kantor & Margolois, 2004 cited in: Cohn, Jia and Larson, 2009, p.224)

It is reported that the Consolidation Standards for Reporting Trials (CONSORT) statement in 1996 was published for bringing about improvement in reporting quality of trials that were randomized and controlled. In addition, the 2001 and 2004 guidelines published by CONSORT have enhanced the transparency in reporting and enabled both editors and readers alike in gaining better comprehension of the design, conduction, and analysis of individual randomized controlled trails. (Cohn, Jia and Larson, 2009, paraphrased) In addition that have been further guideline, 2002es published for appraising and interpreting research and this is reported to include “methods for conducting and assessing systematic literature reviews, the Transparent Reporting of Evaluation of Nonrandomized Designs for Nonrandomized Public Health Studies, the Outbreak Reports and Interventions Studies of Nosocomial InfectionStatement, and Strengthening the Reporting of Observational Studies in Epidemiology, design especially for cohort, case control,and cross-sectional studies (‘Strobe Statement’, 2008).” (Cohn, Jia and Larson, 2009, p.224)

It is reported that Anthony (1996) conducted an assessment of the statistical methods that were utilized in 60 articles and published in a nursing research journal and the report states that there were indiscrepancies found in the analysis of data which included the failure in assessing the data for normal distribution, in matching the data type to the statistical test and in treating paired data (pre- and post-measurements).The report concludes that while the errors were “not necessarily serious, there had been little improvement when compared with the previous decade.” (Cohn, Jia and Larson, 2009, p.224) It is also reported that studies of medical literature have demonstrated “pervasive flaws in statistical analysis. Statistical errors were reported in 85% of 243 studies and 40% of the 154 articles reviewed in psychiatric journals a whopping 52% failed to include information about the statistical method used and 75% used a test that was not correct or a test that had an error in interpretation. In addition, 22% are reported to have drawn conclusions that the data failed to support and 80% did not describe the test used. (Cohn, Jia and Larson, 2009, paraphrased) Mariani and Marubaini (2000) state findings following their review of studies that 81% failed to report an identifiable statistical design.” (Cohn, Jia and Larson, 2009, p.224) No similar studies were found in the field of nursing it is reported.

The work of Brosnan, Eriksen and Lin, (2002) reports that teaching the research process to graduate nursing students takes place in several different ways. One method which is innovative in nature is secondary analysis of large sets of national data and this method is to be such that makes provision of a “meaningful research experience, while saving time and costs.” (Brosnan, Eriksen and Lin, 2002, p.368) However, it is reported that locating data that is useful and that aligns with the clinical interests of students that is easily accessed presents quite a challenge.  It is related that the National Center for Health Statistics (NCHS) both “collects and maintains data” concerning the U.S. population health and additionally provides “summary reports and statistical tables” in addition to making large data sets for public use available. Contained in the data sets are demographic and health status information on U.S. citizens from their birth to their death which are reported as being “appropriate for cross-sectional and sometimes even longitudinal studies.” (Brosnan, Eriksen and Lin, 2002, p.368) Data files include such as: (1) natality data; (2) fetal death data; (3) The National Health and Nutrition Examination Survey; (4) the National Ambulatory Medical Care Survey; and (5) the National Health Interview Survey. (Brosnan, Eriksen and Lin, 2002, p.368) These files are generally available on CD-ROM or diskette and are available for purchase directly from the National Center for Health Statistics (NCHS) for about $20.00 each. Some data sets are available for download directly from the NCHS Web site and the files are reported to be “accompanied by methodological documentation.” (Brosnan, Eriksen and Lin, 2002, p.368)

It is reported that the nursing profession has not historically made use of secondary analysis of large national database or support this approach to any extent in textbooks of published research literature.” (Brosnan, Eriksen and Lin, 2002, p.368) Nursing research textbooks are reported to vary “in the degree to which they treat secondary analysis as a useful research approach.” (Brosnan, Eriksen and Lin, 2002) Other health professionals are reported to use secondary analysis of large state or national database for investigating health issues, it is rare to find nurse authors of such investigations.” (Brosnan, Eriksen and Lin, 2002, p.368) In fact, it is reported that a search of the MedLine database for 1998 and 1999 found that “14 secondary analysis research reports [were] conducted by nurses.” (Brosnan, Eriksen and Lin, 2002, p.368) 13 of these studies were reported to be “secondary analyses of data from relatively small sample research studies.” (Brosnan, Eriksen and Lin, 2002, p.369) Merely one research report utilized a large national data set and during the same two-year period it is reported that “12 research reports, based on large national data sets, were identified as conducted by medical and public health professionals.” (Brosnan, Eriksen and Lin, 2002, p.369) It is clear that nurses are failing to make use of “large repositories of available data. However, these repositories are potentially useful for conducting research concerning myriad health issues relevant to nursing.” (Brosnan, Eriksen and Lin, 2002, p369) Explanations for this oversight are reported to be such as “lack of knowledge concerning the availability of the data, lack of skill in extracting the wanted data, or lack of knowledge in applying the appropriate statistical tests.” (Brosnan, Eriksen and Lin, 2002, p.369)

Data Analysis

Analysis of data collected is often questionable. Kimerling, et al (2010) reports the analyses of members of the military who were screened for sexual trauma including 17580 (80%) women and 108149 (75.8%) men. The study reports that military sexual trauma was reported by 15.1% of the women and 0.7% of the men. Military sexual trauma screen status was significantly related to several demographic characteristics, health care services use, and military service characteristics.” (Kimerling et al, 2010, p.2160) The researchers in the study report that the study results “bear several caveats including that the rate of military sexual trauma and the rate of mental illness reported in this study likely represent conservative estimates because both tend to be underreported. (Kimerling, et al, 2010, paraphrased) While a great percentage of Operation Enduring Freedom and Operation Iraqi Freedom veterans se Veterans Health Administration services, these data do not necessarily generalize to other health care settings.” (Kimberling, et al, 2010, p.2161) Due to analyses being cross-sectional it is reported that the precise timing of military sexual trauma, deployment and the onset of mental health conditions cannot be determined.” (Kimberling, et al, 2010, p.2161) The result is that there can be no conclusions stated in regards to military sexual trauma and the causative of that trauma in regards to the mental health of the veterans in the study. In addition, while frequency and length of deployment which was measured in the reported study is reported as being such that “may serve as proxies for combat exposure.” (Kimberling, et al, 2010, p.2162) It is ultimately reported that more research is needed. research accounting for a broader range of service-related stressors, including both military sexual trauma and combat exposure, is needed.

Data Management

Data management, or indeed lack of proper management of data collected often results in findings in which the data is skewed or simply incorrect. Suris and Lind (2008) report that the VA 2002 national MST surveillance data from approximately 1.7 million VA patients indicated that 22% of women and 1% of men have experienced MST.” Reported is that 22% of women and 1% of men experienced MST and that women are 20 times more likely to be victimized than are men. IN addition, since there are 20 times more men in the military than women and 22% of females and 1% of men screen positive for MST that the numbers of men and women who experienced MST are about equal.  However, since 54% of VA users screening positive are men it is reported likely that there are more male victims of MST than women. These numbers do not make sense because if there are 20 times more men in the military than women and 54% of men screen positive for MST then it is highly unlikely that equal numbers of men and women in the military experience MST but instead it appears likely that many more men than women users of VA have experienced MST.

Rigor

There is reported to have been criticism targeted at qualitative researchers for what is a “perceived failure to demonstrate methodological rigor.” (Ryan-Nicholls and Will, 2009) While qualitative research methods are “accepted as congruent with and relevant to the perspective goals of nursing” there is an ongoing criticism of the methodological rigor or qualitative work.” (Ryan-Nicholls and Will, 2009, p.70) Davidson Reynolds (1971) held that rigor is “the use of logical systems that are shared and accepted by relevant scientists to ensure agreement on the predictions and explanations of the theory. This historical definition highlights the important attributes of rigor arising in a predominantly positivist discourse.” (Ryan-Nicholls and Will, 2009, p.70)

Logic and accepted systems are used in rigorous research according to Ryan-Nicholls and Will (2009) which is reported to provide strength to the arguments made by researchers when demonstrating significance of the findings. Rigor is reported as part of the “iterative, self,-correcting nature of the research.” (Ryan-Nicholls and Will, 2009, p.71) First data are collected and then systematically analyzed and findings presented which “are open to evaluation and replication.” (Ryan-Nicholls and Will, 2009, p.71) Next results that are not correct can be “identified and revised or refuted.” (Ryan-Nicholls and Will, 2009, p.72) It is reported that when quantitative researchers make attempts at applying this definition in determining the rigor of qualitative work that “a dilemma arises”. (Ryan-Nicholls and Will, 2009, p.72) The reported resistance of the positivist to qualitative research is such that extends beyond “the distinction between hard and soft sciences” as positivist practices generate the assumption “that truth can transcend personal bias and opinion.” (Ryan-Nicholls and Will, 2009, p.73) Those supporting quantitative research view qualitative research “as an attack against this tradition.” (Carey, 1989 in: Ryan-Nicholls and Will, 2009, p.73) Positive science and its assault on qualitative research are viewed as “an attempt to impose one version of the truth over the other.” (Ryan-Nicholls and Will, 2009, p.73)

Ethical Considerations

The work of Glenister (1996) entitled “Nursing Research Ethics: Some Problems and Recommended Changes” reports uncertainty in the strategies used to ensure that research in nursing is ethical. It is reported that the codes and guidelines presently available demonstrate limitations and that local research ethics committees(LRECs) appear to be “medically dominated, biased against nursing research, uncertain of the criteria by which they make adjustments, and, in some cases, harmful to those who make submissions.” (Glenister, 1996) In addition, reported is “some evidence of alack of equity in the treatment of nursing research and of research methods.” (Glenister, 1996, p.1)

Richards and Swartz (2002) writes in the work entitled “Ethics of Qualitative Research: Are There Special Issues for Health Services Research” that there is an “increasing volume of qualitative research and articles about qualitative methods…published recently in medical journals. However, compared with the extensive debate in social sciences literature, there has been little consideration in medical journals of the ethical issues surrounding qualitative research.” (Richards and Swartz, 2002, p.135) There are reported to be four potential risks to participants including: (1) anxiety and distress; (2) exploitation, (3) misrepresentation; and (4) identification of the participant in published papers by themselves or others. (Richards and Swartz, 2002, p.135)

It is reported by Richards and Scwartz as follows: “A fundamental ethical requirement of all research is that it is scientifically sound. It should be properly designed and carried out by researchers with adequate levels of expertise and supervision.” (Richards and Swartz, 2002, p.136) Training and supervision issues are specifically “relevant to qualitative research in which the researcher can be regarded as the ‘research instruments’ and in which the researcher will be “often working in isolation.” (Richards and Swartz, 2002, p.136) Ensuring that standards that are agreed upon are met requires ethics committees inclusion or referral to “experienced qualitative researchers when assessing qualitative research proposals.” (Richards and Swartz, 2002, p.137) Furthermore, it must be considered that research in the area of health care is likely to “raise participants’ expectations that help will be forthcoming.” (Richards and Swartz, 2002, p.137) This is stated to be most particularly the case when the participants know that the researcher is a heath practitioner therefore, “explicit ethical guidelines define the duty of follow-up care for participants in quantitative research involving therapeutic interventions, but to date there has been no ethical guidance recognizing the special issues which arise when the researcher is also a health practitioner.” (Richards and Swartz, 2002, p.137) The researcher therefore, must necessarily “be clear about his or her role boundaries, and by ensuring that appropriate information and support are available.” (Richards and Swartz, 2002, p.138) Informed consent is reported as a “prerequisite for all research involving identifiable subjects, except in cases where an ethics committee judges that such consent is not possible and where it is felt that the benefits of the research outweigh the potential harm.” (Richards and Swartz, 2002, p.138) It is stated that an interview study should make a minimum requirement for an interview study of written consent and that the written consent should be obtained from the participant only upon their being informed both orally and in writing about specific issues: (1) the purpose and scope of the study; (2) the types of questions that will be asked; (3) how the results will be sued; (4) the method of anonymization; and (5) the extent to which the responses of participants will be used in reports about the study. (Richards and Swartz, 2002, p.138) In addition, it is reported that participants in the study should be provided a time period to consider their participation and to ask questions of the researcher. (Richards and Swartz, 2002, paraphrased) Qualitative research is such that “unexpected themes can arise during the analysis; therefore, at the time of the interviews, the potential uses of the data are not always clear to the research team.” (Richards and Swartz, 2002, p.139) Additionally, the researchers may desire that they are able to access interviews at a later date and therefore may archive interviews. Participants should be fully informed of this and given the opportunity to withdraw their consent for use of their data. If they so desire the participants should be allowed to do so. (Richards and Swartz, 2002, p.139) Two approaches exist for ensuring that adequate consent is acquired in qualitative studies: (1) participants can be asked to give very general consent when the study begins; or (2) researchers can treat consent as an on-going process rather than a one-off event. (Richards and Swartz, 2002, p.139)

Conclusions

The literature reviewed in this study has demonstrated questionable approaches in qualitative research and most specifically in regards to data analysis methods and the findings rendered in some qualitative studies. Even in quantitative studies data analysis has been found, based on the literature reviewed to be questionable. There are many considerations in nursing research that must be examined prior to ensuing in a research initiative. Nursing professionals who engage in research should necessarily interact with mentors and more experienced nursing professionals prior to engaging in research to ensure that they meet the necessary standards and guidelines in order to meet the standards of rigor that are required in nursing research.

Bibliography

Barrett, RD (1994) Using the 1990 US Census for Research. Thousand Oaks, CA. Retrieved from: http://books.google.com/books?id=7Az6T-b_V_QC&pg=PA254&lpg=PA254&dq=nursing+research:+secondary+analysis&source=bl&ots=4ybNM6vKlR&sig=BdnANa2yHNSucu3O7bKRqo9dNcQ&hl=en&ei=pMRjTo-pKKjV0QGjzqypCg&sa=X&oi=book_result&ct=result&resnum=8&ved=0CHYQ6AEwBw#v=onepage&q=nursing%20research%3A%20secondary%20analysis&f=false

Bibb, Sandra C. Garmon (2007) Issues Associated with Secondary Analysis of Population Health Data. Uniformed Services University of the Health Sciences. Retrieved from: http://digitalcommons.unl.edu/cgi/viewcontent.cgi?article=1001&context=usuhs&sei-redir=1#search=%22nursing%20research%3A%20secondary%20analysis%22

Bierman, A., & Bubolz, T. (2003, October). Secondary analysis of large survey databases. In M. B. Max, J. Lynn (Eds.), Interactive textbook on clinical symptom research (chap 20, pp. 2–3) Available at: http://symptomresearch.nih.gov/chapter_20/index.htm.

Brosnan, CA, Eriksen, LR and Lin, 2002, Yu-Feng (2002) Teaching Nursing Research Using Large Data Sets. Nursing Education Journal. Educational Innovations. August 2002, Vol. 41, No. 8.

Clamp, CG, Gough, S and Land, L (2004) Resources for Nursing Research. SAGE 2004.

Cohn, EG, Jia, H and Larson, E (2009) Evaluation of Statistical Approaches in Quantitative Nursing Research. Clinical Nursing Research August 2009. 223-241. 2009 SAGE Publications.

Du Plessis, E and Human, SP (2009) Reflecting on Meaningful Research: A Qualitative Secondary Analysis. Curationis, 2009 Sep, 32(3) 72-9. PUBMED. Retrieved from: http://www.ncbi.nlm.nih.gov/pubmed/20225746

Glenister, D. (1997) Nursing Research Ethics: Some Problems and Recommended Changes. Journal of Research in Nursing, May 1996. Vol. 1 No.3 184-190. Retrieved from: http://jrn.sagepub.com/content/1/3/184.abstract

Haskell, S.G. et al (2010) Gender Differences in Rates of Depression, PTSD, Pain, Obesity and Military Sexual Trauma Among Connecticut War Veterans of Iraq and Afghanistan. Journal of Women’s Health. Vol. 18 No. 2. Mary Ann Liebert, Inc.

Heaton, Janet (1998) Social Research Update. Issue 22, Autumn 1998. Sociology of Surrey. Retrieved from: http://sru.soc.surrey.ac.uk/SRU22.html

Himmelfarb, N., Yaeger, D. and Mintz, J. (2006) Posttraumatic Stress Disorder in Female Veterans with Military Civilian Sexual Trauma. Journal of Traumatic Stress. Vol. 19, No.6. December 2006. Wiley Interscience.

Jacobson, AF, Hamilton, P, and Galloway, J (1993) Obtaining and Evaluating Data Sets for Secondary Analysis in Nursing Research. West J Nurs Res 15 Aug(4): 483-94. PUBMED. Retrieved from: http://www.ncbi.nlm.nih.gov/pubmed/8356839

Jeffs, L et al (2006) Building Capacity for Nursing Research. Can Nurse 2006 Oct;102(8)30-2. PUBMED. Retrieved from: http://www.ncbi.nlm.nih.gov/pubmed/17094366

Kimerling, R, et al (2007) The Veterans Health Administration and Military Sexual Trauma. American Journal of Public Health. Dec 2007, Vol 97, No. 12.

Kimerling, R. et al (2010) Military-Related Sexual Trauma Among Veterans Health Administration Patients Returning from Afghanistan and Iraq. American Journal of Public Health Aug 2010, Vol 100, No. 8.

Kniepp, SM and Yarandi (2002) Complex Sampling Designs and Statistical Issues in Secondary Analysis. Western Journal of Nursing Research 24(5) 552-66.

Mariani, L., & Marubaini, E. (2000). Content and quality of currently published phase II cancer trials. Journal of Clinical Oncology, 18, 429-436.

Orsi, AJ, et al (1000) Conceptual and Technical Considerations When Combining Large Data Sets. Western Journal of Nursing Research 21(2) 130-42.

Polit, DF, Beck, CT (2004) Nursing Research: Generating and Assessing Evidence for Nursing Practice. Lippincott Williams and Wilkins.

Richards, HM and Schwartz, LJ (2002) Ethics of Qualitative Research: Are There Special Issues for Health Services Research? Family Practice. Vol. 19 No. 2 Oxford University Press 2002. Retrieved from: http://fampra.oxfordjournals.org/content/19/2/135.full.pdf

Riedel, M. (2000) Research Strategies for Secondary Data: a Perspective for Criminology and Criminal Justice. Thousands Oaks, CA.

Ryan-Nicholls, D and Will, CI (2009) Rigour in Qualitative Research: Mechanisms for Control. Nurse Researcher 2009, 16, 3.

Smaldone, A. M., & Connor, J. A. (2003). The use of large administrative data sets in nursing research. Applied Nursing Research, 16, 205– 207.

Stommel, M and Willis (2004) Clinical Research: Concepts and Principles for Advanced Practice Nurses. Lippincott Williams and Wilkins, 2004.

Suris, A. and Lind, L (2008) Military Sexual Trauma: A Review of Prevalence and Associated Health Consequences in Veterans. Trauma, Violence & Abuse. Vol. 8 No. 4, October 2008. Sage Publications.