Research Tools and their Effectiveness - Essay Example

Research Tools and their Effectiveness In this study, the design used is monitoring the effects of the training in the accuracy red hot system. 8 radiographers are observed for their sensitivity, specificity and accuracy of its use and then monitored again after training and how they respond to the three effects; sensitivity, specificity and accuracy. The research under scrutiny is conducted in a sensitive environment where they are being tested using community health services in two different scenarios. It has two settings; one is before training while the other is after training. This is to take note of the level of radiograph each scenario can administer with the community physiotherapists at heart. The research took more than 26 weeks to complete. With the sensitivity of the matter, training them for only 8 weeks doesn’t make a lot of sense in that they will not have experienced the necessary knowledge which they can use with ease to determine whether to report as casualty or not. Further, the issue of monitoring the sensitivity, specificity and accuracy is not much encouraged because, if you monitor them, they will tend to be accurate and may not sway away from normality due to your presence. Restrictions For a research to have the required scientific vigor homogeneity of data is required and this can only be obtained if some certain restrictions are imposed on the participants (Agnes, A 2007). This may include: Age, gender, occupation, health status, earning and many others depending entirely on the objective of the study. For example, for one to take part in this study, the researcher ensured that, he/she had at least some experience in radiography which he himself knew and another which had brought him to that health facility to assist in the provision of the service. However, they were not required to have any formal training in radiography. Ethical Considerations In any scientific research which involves human participants, freedom to take part in it is a priority as you cannot force one to get involved without his/her consent (Mailu, S. 2004). Participants have to be given the opportunity to say what they want based on the restrictions above. In this study, all the participants are not provided with the necessary information and a verbal consent is not sought from each of them. Mainly a consent form is signed by him/her stating that, all that is involved in the study will be followed to the later voluntarily (Charles, A 2000). The protection of human participant’s information is key for any research to be successful (Bailey, 2005). Also, on signing the consent form, the research and the participant sign an agreement that, the participant’s information will be only used for the intended research work and nothing else. According to human conventions, patients qualify to be called vulnerable groups and thus, it is important to protect them under all circumstances. This is not mentioned anywhere in this study as it should have happened in an ideal research study. Sampling Strategy A sample in mostly used in inferential statistics to draw information about a certain population. Sampling is done in many ways (Molo, S 2006); there is simple random sampling where each element in the population has an equal chance of being selected, systematic sampling where elements are chosen in a certain order, cluster sampling and stratified sampling, where the population is put into strata each with a certain type of population entities, which are highly recommended in many research works due to its coverage (Oloo, 2003). According to Barnes (2007), it is highly recommended in many instances due to the following: It is cheap to carry out in terms of resources, it takes a shorter while than a complete count, and it has a higher scope in terms of the results gotten from the data collected among many others. Sample selection mainly depends on the whole population, the level of significance (mostly 95%) and the amount of variations as far as the traits are concerned. It should be an un-biased representative of the whole population and in most cases; stratified sampling is used due to the heterogeneity of populations so as to make it homogeneous which is good for any research work. Further, the data should always free from bias, the gender of the participants should be taken into consideration. Men and women should have unbiased representative. But in most cases, the population dictates the representation. This research concentrated on radiographers only and out of the 8 who took in both the scenarios, no mention of their gender status. The sample For any researcher to attain tangible results, the sample collected should represent well the population of the elements (Barnes, 2004); in this case radiographer. In this case, the number 8 is not well representative since it seems a small number combared to the number in existence. A sample size should always be directed by the bound on the error of estimation as well as the level of significance. Also, the sample should not be biased in terms of either male or female (should be gender sensitive). This case doesn’t talk of this and is open to bias. In selection, stratified sampling should have been used since all of them are not of the same cadre and that some difference does exist. For example, their work experience has a lot to do with the way he/she will react before and after training. Further, his/her background will have a lot, so, all this combined makes ones to consider using another method and not randomly. Data Collection Methods The background information part aimed at collecting data on the characteristics of the radiographers which included age, ethnicity, gender, marital status, employment and income (Richmond, 2005). This is reliable and valid in that, it is provided by someone meaning he/he was not coerced and that his background is not definite, actually randomization took place. In this case, this is not done and thus exposes the work into some questions In any research, the methodology of conducting it determines the validity and reliability of the data gotten from the research tool (Mallinson, Sara 2003). This research tools ranges from questionnaires which may either be self administered or posted based on the study, observing the participant’s behaviours, using phone, using websites among many others. In this case, monitoring is used. This does not produce all the required information since some cannot be gotten by mere observation and the researcher needs to document others from either files or past knowledgeable records on how the individual has been performing before and whether the training will have any impact whatsoever. In additions, for the same to be reliable and valid, the research tool is to be tested first of all using by conducting a pilot survey (Anthony, Glenda 1996). Further, the tool should be exposed to the require vigor by, if possible, convening a panel of experts in that field to think aloud on the issues to be asked and to come up with concrete and not ambiguous questions for the participants (Amos, V 2004). In this case, the tool is not exposed to high level test-retest to deem it fit for the study. In this study, the researcher colleted the data using an observation (monitoring) and face to face interview while recording the responses from the participant in the first instance. This method is highly recommended since the participant has the will to devolve all the information he/she is being asked for and the researcher will observe for himself if that s true or not. Further, the researcher can ask for more clarifications if he/she did not get the required information. In case the questions are open, he/she can frame the answer in such a manner that, the final coding will require very little at the end. Further, in order to collect some more tangible data, the researcher should have considered using other methods e.g. questioning their seniors of the way they behave and comparing that with other ways. Analysis Data collected from the radiographers was first of all coded and then entered into a computer and quantitative results produced. Coding data is mostly done from open questions where the participant is not restricted in the way to answer them. Coding originates from the highly quoted answer to the question. These codes are mainly numerical which are then entered and analyzed as numeric (Otieno, A 2003). In data analysis, quantitative results are highly valuable in that, they say exactly the number of people in support of a certain way of doing things. This are represented as pie charts and tables which capture the reader eyes. The all radiographers’ results are good as they tell overall how they view their learning modes and how they can apply what they have been trained on. The use of t-test is not the best since the observations cannot be assumed to be normally distributed to warrant a t-test as in the case. Mann-Whitney test should have been used and not a t-test. In Mann-Whitney, no assumption of normality. Likewise, the three monitoring factors cannot be quantified and thus the use of the following tables cannot be wished away, instead of this, the following should have been used. The research findings in this case are not valid and reliable since the tools used in the study don’t represent all the well known in that field of data collecting. In analysis, he/she should have considered using the following if tables as an example. 1. What is your age? Age Frequency 20-30   31-40   41-50   50+   2. Education level Education level Frequency Primary   High School   College   University   Other   3. Work experience Experience (years) Frequency 1-10 years   11-15 years   16-25 years   25+ years   4. The effect o training Will training help Frequency Yes   No   And many others. Further, he/she should have considered using charts and graphs since this provide a bird’s eye view of what is happening. Generalazability This data gotten form this analysis cannot be used to generalize the results. First is because o the sample collected which is not representative and which is small compared to the number of radiographers in existence. Secondly, the results are narrow since they are conducted at 95 percent CI on a very sensitive matter and if possible 99 percent should be used in this case. Figures don’t lie and in this case, the statistical test developed does not show any significance difference in the two scenarios; before and after training, thus the conclusion that, the two are different are baseless and cannot be trusted at 95 percent. Further, the study points at where radiographers make most mistakes in although they are not acted upon during the study. Conclusion In conclusion, although the research is done at some level of scientific rigor, in terms of the sample and analysis, it has a lot left to be desired. According to Peter, (2007), the sample of any study should be fully representative and should attain the required objective. James (2003) sums up this by saying that, any research which allows a short while before the transition of the research scenarios, then it is an hanging research work. Reference: Agnes, A (2007). Carrying out research using human participants: What you should do and not do. Ohio: Akron University Press Anthony, Glenda (1996). ‘Active learning in a constructivist framework’. Educational studies in Mathematics. 31 (4), 1996 pp. 349-369 Bailey, (2005). Research Tools and their Effectiveness. Baltimore: John Hopkins University Barnes, (2004). Research Ethics: Sampling Techniques and how they should be used to attain rigor in the study. Nairobi: General printers Barnes (2007). The importance of Sampling: Why it is recommended in many organizations. Nairobi: General printers Charles, A (2000). Participating in research: Should it be voluntary or mandatory? Nairobi: Government Press James (2003). Planning a successful research work. Ohio: Akron University Press Oloo, (2003). Conducting Ethical Research Work Using Statistics. London: Longhorn Otieno, A (2003). Data coding: The rules used in setting up the codes. London: Longhorn Mailu, S. (2004). Statistics Application in research: How the findings can be used to make decisions. Nairobi: Government Printers Mallinson, Sara (2003). ‘Listening to respondents: A qualitative assessment of the Short-Form 36 Health Status Questionnaire.’ Social Science & Medicine, Vol. 54, pp. 11-21 Molo, S (2006). Statistical Analysis. London: Penguin Peter, (2007). Scientific Research requirements. Melbourne: penguin Richmond, (2005). Data reliability and validity: Collecting data to attain both. Nairobi: General Printers Read More