There are various techniques used by researchers to carry out quantitative research. These may be some form of experiment, interview or various survey forms such as attitude measures or questionnaires, (Eachus, 2003).
Methodology
Various parts of the Cormack framework cover the methodology, it asks if the subjects are clearly identified, if the approach to the sample selection is clearly stated, if the sample size is clearly stated and about the data collection, (Cormack, 2000).
For this study Birtwistle et al carried out a cross-sectional postal survey using an anonymous questionnaire. They say this is one of the best ways to gather large amounts of information about a research area when there is a scarcity of information, (Birtwistle et al, 2002).
As this was the first research to be carried out on this subject the authors wanted a large amount of information as a starting block for the future. Their use of a postal questionnaire survey has many benefits, it is usually the cheapest form, anonymous and postal research tends to provide a truer reading as respondents are more likely to report their true feelings of criticism or satisfaction as their anonymity is not compromised, (Crow, Gage, Hampson, Hart, Kimber, Storey, Thomas, 2002).
We are told that the survey was carried out in the districts of East Dorset, West Dorset, Isle-of-Wight, Southampton and South-West Hampshire. There is no rationale or explanation offered for why these areas were chosen, (Birtwistle et al, 2002).
As the research title does not specifically say district nurses in the South of England then the whole population of district nurses in Britain should have been given an equal chance at being selected. As only a small section of the country was selected, for whatever reason, then this is a sampling bias, (Eachus, 2003).
From the target districts any qualified nurse who worked in the community was put forward for selection but excluded Health Visitors, Community Midwives, Practice Nurses, School Nurses, Community Psychiatric Nurses and Occupational Health Nurses. Current employment records were used to identify the possible candidates. To maintain confidentiality the questionnaires were not posted directly by the researchers but by the district coordinators. Over a period of eight months throughout 2000 a total of 522 questionnaires were sent out followed up by reminders at four and seven weeks. The method of data collection was through a questionnaire. The questionnaire was based on one that had previously been developed and used by two of the authors, Payne and Kendrick. This final questionnaire was then pilot tested. It contained sections on demographic information, interest in and education on bereavement, the nurses’ view on bereavement care and their perceived role in this care, the normal care provided to bereaved patients, and information about the general practice the nurse worked from. As briefly mentioned previously the questions asked the nurse to tick a box from a choice of two or more possible answers. The section on the perceived role of the nurse in bereavement care required a response on a five point lickert scale from strongly agree to strongly disagree. Nine of the questions also allowed a section for the nurse to add additional information, (Birtwistle et al, 2002).
Using a Lickert scale is one of the most popular forms of scale measurement, they are named after the psychologist Rensis Lickert, (Polit and Hungler, 1999).
Lickert scales are generally based on a five point scale of a person’s opinion; this would normally be set out as ‘strongly agree’, ‘agree’, ‘undecided’, ‘disagree’ and ‘strongly disagree’. However this format is not the only type in use today, other variations are used. For example a scale may be one line with ‘strongly agree’ written at one end and ‘strongly disagree’ written at the other end or a line with -5 written at one end and +5 at the other, (appendix 1), (Cormack, 2000).
There are many advantages in using questionnaires, if as in the case of Birtwistle et al’s study it is a postal questionnaire then large numbers of people can be reached over a larger area at a lower cost than most other methods. Due to the fact that they are structured and predetermined they carry a reasonable degree of reliability and they also allow respondents to answer in their own time at their convenience, (Parahoo, 1997).
Birtwistle et al clearly state that they used the SPSS or Statistical Package for the Social Sciences version 10 for their data entry and analysis. Chi-squared tests were used to identify variables associated with reported roles of the district nurse, levels of interest in bereavement, feeling informed and the likelihood of post bereavement follow-ups. The Chi-square test is used when there are categories of data concerning the proportions of cases falling into the various categories. It is used to test the significance of different proportions, (Polit and Hungler, 1999).
The key variables used for analysis were age, pre and post-registration education, employment status, personal loss, level of academic achievement, district, professional position and general practice activities in relation to death. Ordinal data such as age were analysed using cross-tabulations. Additional comments were grouped into themes with examples. All direct quotes are identified with a code letter to indicate the district and a number to identify the respondent, (Birtwistle et al, 2002).
Appropriateness of Approach and Methodology
When selecting a research design the methods used may be the personal preferences of the researcher. Quantitative is commonly selected over qualitative due to its scientific nature and surveys conducted using questionnaires may well be selected over interviews due to factors such as cost and time. However short-cuts should not be made if the best knowledge is to be gained. It is best to use the most appropriate strategies if the results are to be valid and reliable, (Parahoo, 1997).
Parahoo (1997) describes how the survey method is appropriate for descriptive and correlational studies, used in order to glean large amounts of data from large representative samples. In correlational studies researchers generally collect demographic details such as age, gender and education. They seek to establish the links between these and other characteristics of the respondents such as their beliefs and behavior. This is why surveys are ideal methods of data collection for these types of studies, (Parahoo, 1997).
Brown, Tanner and Padrick (1984) as cited in Parahoo 1997 found that questionnaires were the most common research instrument used when they analysed a sample of 137 studies. The small amount of evidence available would suggest that the questionnaire is the most widely used method of data collection in nursing research. Questionnaires can be efficient in providing data on the attributes of staff and clients and can be used in the evaluation of the needs of staff and clients. They are also used to collect information on knowledge, beliefs, attitudes and opinions of staff and clients. The information on factors such as attitudes, beliefs and knowledge is very useful in research as these can influence the practice of health care professionals, (Parahoo, 1997).
The Cormack (2000) framework asks if the strengths and weaknesses of the approach chosen are stated. Birtwistle et al have described the strengths and weaknesses of their study design. They describe how the postal questionnaire is an effective way of gaining large amounts of data but they admit this data may lack depth. They were also aware that the respondents may try to give the answers they felt were correct rather than what they really thought or felt. Birtwistle et al tried to minimize this problem by ensuring the questionnaires were anonymous, (Birtwistle et al, 2002).
The response rate to Birtwistle et al’s postal survey was 62%, compared to other survey response rates this appears to be quite reasonable; however they do state that with this response they are still lacking the opinions of 38%. Research has found that anonymity is an efficient way of helping to increase the response rate particularly if the topic is sensitive, (Cormack, 2000).
Cormack’s framework also asks whether the sample selection approach is congruent with the method used. Birtwistle et al have chosen a representative sample of the target population. This sample was large (522) which is ideal for a quantitative questionnaire study. They could have used a much larger number of people however time, funding and resources may be constraints to using a larger sample, it should be remembered that sampling beyond a certain point will achieve nothing and is wasteful, (Eachus, 2003).
The use of a survey questionnaire for a relatively large quantitative study seems to be popular. Having carried out a search for other such studies the author found that many quantitative studies used this method.
One such study was carried out by Llahana, Poulton and Coates; it was published in the Journal of Advanced Nursing. A survey design was used with a questionnaire consisting of qualitative and quantitative approaches. This survey was carried out on 80 people selected from a directory of appropriately qualified nurses. This was an anonymous postal questionnaire with confidentiality assured. The questionnaire asked for demographic data such as gender and qualifications and also asked questions about their employment with answers given on a five point Lickert scale. The data collected was analysed using SPSS. The response rate was 66 people (82.5%), (Llahana, Poulton and Coates, 2001).
Evidence Based Practice
The purpose of nursing research is to increase and improve the knowledge that practice is based upon. Nursing research is still relatively new and a good deal of today’s practice is based on tradition, beliefs and intuition rather than research evidence. Nursing research is an attempt to change this practice by providing the nursing profession with a base that can be defended on the grounds of scientifically established knowledge. It also aims to provide a basis for setting standards and giving quality assurance, to increase the cost-effectiveness of nursing practice, to give evidence of the strengths and weaknesses in nursing, to earn and support a professional status in nursing and to give the term evidence-based practice scientific credibility, (Cormack, 2000).
Evidence-based nursing requires practitioners to base their practice on the best evidence available from current research rather than relying on intuition, trial and error and personal experience, the source of this evidence is from quality research, (Cormack, 2000).
Not all research is high quality, so evidence-based practice involves the gathering and evaluation of research. The information used to create evidence-based practice will depend on the suitability of the evidence for the issue in question. Not only research can be used for this but also clinical guidelines, expert opinions, benchmark standards and consensus views. However primary research findings are predominantly considered as the highest form of evidence. Other information sources such as clinical guidelines tend to be derived from research and are therefore considered a form of secondary evidence. Systematic reviews are a valuable method of evaluating large volumes of evidence, if carried out correctly they can provide a reliable summary of the best evidence available. Nurses need to be able to evaluate these systematic reviews to determine its relevance for individual client care, (Hewit-Taylor, 2002).
The Cormack (2000) framework asks if the clinical implications are discussed in the study. Birtwistle et al say that district nurses are ideally placed to carry out bereavement support however they conclude that with the lack of relevant research and appropriate training the suitability of district nurses is debatable. They also tell us their findings have important implications on the education and training of district nurses in the field of palliative and bereavement care. Their results indicated that few nurses had received bereavement knowledge or training from their pre-registration education. Post-registration training tended to be specific to nurses with an interest in bereavement. Cormack’s framework also asks if the recommendations in the study suggest further areas for research. Birtwistle et al’s recommendations suggest the need for further research in this area due to the lack of any previous, they particularly recommend a qualitative study. To investigate the quality of service provided it is recommended that a national study be carried out in order to develop guidelines for best practice. It is also suggested that the benefit of an information pack on bereavement for clients should be evaluated. Their last recommendation is the need for a reassessment of both pre-registration and district nurse training, specifically in bereavement support and palliative care. This would have important implications for the care received by patients and families of the terminally ill in the community. Further to the study Birtwistle et al have now carried out a qualitative research study with district nurses using a series of in-depth audio-taped interviews. This is currently being analysed but has not been reported as yet, (Birtwistle et al, 2002).
It appears that this study is a good foundation for future research to be carried out in this domain. It will have served as good underpinning knowledge for Birtwistle et al to carry out their qualitative study by bringing the most pressing issues to the fore.
Appendix Page
Appendix 1:
Lickert Scale
- Nursing models improve care SA A U D SD
- Nursing models involve too much paper work SA A U D SD
- Nursing models work well in practice SA A U D SD
Taken from Cormack (2000) Page 281.
References
Birtwistle, J. Payne, S. Smith, P. and Kendrick, T. (2002). The Role of the District Nurse in Bereavement Support. Journal of Advanced Nursing. Vol. 38, No. 5, pp. 467-478
Cormack, D. (ed). (2000). The Research Process in Nursing. 4th ed. Oxford: Blackwell Sciences Limited. pp.7, 17, 106, 281, 309.
Crow, R., Gage, H., Hampson, S., Hart, J., Kimber, A., Storey, L. and Thomas, H. 2002. The Measurement of Satisfaction with Healthcare: implications for practice from a systematic review of the literature. [Online] Health Technology Assessment. Available from: [Accessed 20 Sept 2003]
Eachus, P. Dr., 2003. Research Methods for Health Sciences. [Online]. Salford: University of Salford. Available from: [Accessed 2 Sept 2003]
Hewitt-Taylor, J. (2002). Evidence Based Practice. Nursing Standard. Vol. 17, No. 14 & 15, pp.47-52 and 54-55.
Journal of Advanced Nursing. 2003a. Journal of Advanced Nursing. [Online], Available from: [Accessed 4 Sept 2003]
Journal of Advanced Nursing. 2003b. Journal of Advanced Nursing. [Online], Available from: [Accessed 4 Sept 2003]
Llahana, S., Poulton, B. and Coates, V., 2001. The paediatric diabetes specialist nurse and diabetes education in childhood. Journal of Advanced Nursing. [Online].33(3), Available from: [Accessed 22 Sept 2003]
Parahoo, K. (1997). Nursing Research: Principles, Process and Issues. Hampshire: Palgrave Macmillan. pp. 81, 82, 143, 148, 149, 248, 262, 263.
Polit, D. and Hungler, B. (1999). Nursing Research: Principles and Methods. 6th ed. Philadelphia: Lippincott. pp.338 and 487.