Abstract Research methods in social sciences have two basic categories; qualitative and quantitative methods. Quantitative research involves asking people's opinions in a structured manner in order to produce hard facts and statistics. In contrast, qualitative research is an explanatory research with the main focus being to gain insights into the helps or problems to develop a hypothesis for potential quantitative research (Cohen et al., 2010). Quantitative research is used to quantify defined variables such as opinions and attitudes, and generalizing results from a larger sample. They use measurable data in formulating facts and unearthing research patterns. Qualitative data collection methods take the form of surveys and include face to face interviews, systematic observations, phone interviews, longitudinal surveys and mobile surveys. In this paper, there are academic critique of the article "The effects of teacher training on new instructional behavior in reading comprehension" by de Jager, Reezigst and Creemers (2002). Also, a Research paper on, “The effects of hospital palliative care programs on the seriously ill patients’ family members” by Gelfman, Meier and Morrison (2008). The evaluation systematically focuses on specific aspects of the articles in terms of process and validity of research methods and results. The papers have tried to come up with a cohesive explanation that not only elaborates on the researches particulars but also formulates clear interpretation of the researchers simultaneously. As per the suggestion, the sample sizes and the methods of selecting the subjects for use only allow for internal validity of the study thereby diminishing chances of result generalization to the immediate community. Variable Specification and Data Measurement De Jager et al. (2002) in their study outlines a quasi-experimental research design involving three sample groups; two experimental and one control drawn from an immediate population of 83 primary school teachers in Northern Netherlands. The purpose of the research was to reiterate the need for the education to be constructive focused, student oriented with active teacher's participation in facilitating and guiding the pupil to their understanding. The research focused on exploring possible means of reconciling educational theory and practice. The foundation of any research is the definition of an independent and dependent variables. The research independent variables are the training of the two experimental groups; the Direct Instructions (DI) and Cognitive Apprenticeship groups (CA). The inclusion of teaching methods is the dependent variable in order to develop metacognitive reading skills. Of the 83 individuals in the community, three sample groups were created from 20 volunteer subjects. The first experimental group composed of eight teachers' in-service in CA, second consisting of 5 teachers in-service in DI with a control group of 7 teachers. The data are collected through direct interview using a structured questionnaire for first hand information. The respondent offered guidance to the respondents in filling the questionnaire. Like any other quasi-experimental research, the research included some methods or tools for evaluating and estimating the causal effect of the autonomous variables on the explained variable. In terms of the evaluation design, the authors developed a survey style observation tool that was used for four times for each experimental group and only twice for the control group. The instrument fell in the category of non-parametric testing specifically designed for the measurement of these groups. An equivalence of the groups before treatment was tested with the Kruskal Wallis One Way analysis for independent groups- an ordinal ranking test. In the evaluation of generated data from low and high inference portions of the observation tool, “The Mann-Whitney U test” was utilized. The observation tool composed of two sections; low reference and high inference sections. The low inference section consisted of a two minute duration score checklist. It indicated the activities were taking place at each interval in quantitative form. The high inference evaluation structured in an ordinal scale of one to five. The values indicated the level of satisfaction and the willingness to change.
The nonparametric observation tool creates the problem of external validity by ignoring the normality and score distribution. The inter rater reliability is measured on a scale of -1 to +1. As the score approaches +1 its reliability increases and vice versa. The study inter rater reliability was quite high with a scale of 0.81. Gelfman et al. (2008) aimed at examining if hospital palliative care consultations improved the quality of care. The study uses telephone interviews on a sample of 190 members with only 149 successful interviews. In order for a subject to qualify, the deceased must have had a hospital admission of greater than 10 days, consultation for palliative care 3 days to death or the receiving of usual care. The consulted family members must have had their loved one passed away between the months of April and December 2005. The three months after death grounded on the experimental knowledge that the family members appear to have minimal distress three months after death (Kaiser Health News, 2010). More so, the family recall of hospital events was still reliable and valid for six months following the death of their loved one. The data were more in the form of descriptive statistics. The emotional and spiritual needs of the family were measured using the ordinal scale of one to five of “strongly disagrees” to “agree strongly”. In assessing the extent of meeting the emotional and spiritual needs under general and palliative care, the research used a nominal scale with a “Yes” or “No” response. The study utilized an ordinal scale of “fair”, “not fair” and “no comment” in rank of 1, 2 and 3 respectively in the evaluation of self efficacy. In estimating the level of confidence with the domain, the respondents were required to give a “Yes” or “No” answer (Kaiser Health News, 2010). Descriptive statistics was prevalent in order to condense the large amount of data in a usable form. The data was condensed into percentages in order to represent the subgroup of the total. A large number of members call for statistics validation.Variations within the Data: De Jager et al. (2002) study is shown to be quasi-experimental with pre-test and post test control design. In the study, the manipulation of independent variables is done in such a way that interservice of both control groups involved a methodological approach. On top, they are autonomously treated while both dealt with metacognitive skills, the skills impacted in very different means. The control group received no training or technical aids, and was not influenced by the researchers beyond the fact that, during the two lessons in a particular school year, with each member observed during the delivery of a lesson on reading comprehension. This withholding there is quite a large number of extraneous variables not taken into account. The result for this on business process is that the policy formulation based on this will be vague and unreliable. The virtue that the research never narrowed down to the subject of choice eliminates the confidentiality of the whole process. More so, the subjects were not subjected to the same condition makes it hard to be relied on for tastes and preferences in consumer evaluation. The study includes unequal groups by incorporating the students' performance. In business techniques, elements of the same group are the only one involved in the case of individual discipline study. The result is to help generalize the entire population results and findings as per the sample. Gelfman et al. (2008) found a correlation between developments in family-centred results and the palliate care service. There was a positive response from family members of the deceased who received palliative care services. The family members' confidence in their own self efficacy skills was observed to have improved in the involvement of taking care of their ill loved ones with 68% level of confidence. The results demonstrated the reliance of the first hand information in deriving a conclusion as the previous study on the same subject had made. The same should prevail in business research where the interviewer identifies out the affected groups and mark them out for an interview. In carrying out business researches, there is a need to include both quantitative and qualitative research methods. The move ensures that analysis and conclusion utilize both descriptive and prescriptive data. There will also be an improvement in data quality as qualitative research forms the basis for quantitative research. Concluding the Results Findings of the de Jager et al. (2002) prove their hypothesis to be correct. The authors attribute the relative change in institutional strategy within the experimental groups to the independent variables treatment. The statistical results of the samples were represented in separate sections and tables. As such, it is evident that the researchers intentionally avoided comparing the DI and CA instructions models. The unequal treatment of subjects creates an error in the study resulting in variations. The resulting two different charts combination into a single chart failed. The researchers should have subjected all the subjects to the same conditions in order to reflect a particular institutional method. The result would be high inference. Wholesomely, the whole article lacks in terms of external validity. As per the authors finding there is a need to amend the research methods employed. The recommendation will help improve the result conclusiveness based on the data. The study offers no generalized results as per the expectations and only show how the training in alternative instructional methods can affect a regular teacher teaching methods. The selection of sample groups was poorly made failing to consider the immediate group. As a result, the result findings are specific to the individuals studied. The regular teachers are taken to be equal forcing the authors to be concluding of possible representation allowing them to recommend for replication of their work. Apart from that, the result fails to show any real difference between the sample and the control groups. The research, unlike other quantitative researches, shows much subjectivity other than objectivity. The authors on one hand clearly show that very experienced educators on the profession will not readily accept changes in their instructional methods. On the hand, they select a representative group with statistically 95% of its objects of study in the study having not less than 10 years of experience. The intention of the action is not clear and as such bringing in subjectivity on the researchers' side. In Gelfman et al. (2008), the research process supports the initial research question of the researchers. Of particular concern, hospital based palliative care programs immensely increase family satisfaction with care and results. The findings facilitate the authors' conclusion and policy recommendation on the development in hospital palliative care programs and enhance physician trainings. Numerous errors may have emanated from the research procedures and analysis. Firstly, non English speaking families were excluded from the study creating a drawback on the generalization regarding them based on the English speaking. The research could have hired a person to contact the non English speaking groups for binding issues. Also, the research only considered families who had next of kin listed in the medical record with a valid phone number. The recall of family members was subject to bias. As a result, the researchers defined a specific time frame of 200 days after the passing away of a loved one. Another limitation was the non inclusion of families and patients with a short stay in the hospital before death or those discharged alive. The researcher opted for a large hospital in New York City. The action creates the question of pertinence of results in other populations. There was need to include different settings and with a varied group in order to avoid biases based on the pressure available. The original study group was not randomly selected or following a defined formula rather it was a list of subjects. The result may have led to extraneous variables being included in the research selection process. More so the use of telephone in the interview created a limitation in the survey. The result was validation of responses without further testing. There was no personal contact which could have allowed for additional information to the researcher. In the future, the researchers should incorporate various methods of research to eliminate bias created by only one. The element of home care and treatment was greatly ignored. In most cases, the patients are accorded home care and treatment before being taken to hospital for admission and personalized care. In a nutshell, the paper is an efficient and effective critique of the two quantitative research papers. The critique has used five criteria in analyzing the quantitative researches. The result is an in-depth understanding of quantitative research methods for business.
Cohen, L., Manion, L., & Morrison, K. (2010). Research Methods in Education (5th ed.). New York: Routledge/Falmer.
De Jager, B., Reezigt, G.J., & Creemers, B. (2002). The effects of teacher training on new instructional behavior in reading comprehension. Teaching and Teacher Education, 18, 7, 831-842.
Gelfman, Meier, & Morrison (2008). The effects of hospital palliative care programs on the seriously ill patients family members
Kaiser Health News. (2010). Palliative care helps improve quality, cut end-of-life care costs. Retrieved from http:www.kaiserhealthnews.org/Daily-Reports/2010/March /01/End-of-life-issues.aspx