Research method Essay
This research will examine the importance and position of employees in terms of Job Satisfaction and career growth in Jamaica and assess the way in which the needs of the same are met by the banking industry. The approach will be based on proven methodologies applied to the market, starting with an in-depth literature review followed by primary research. This research will concentrate on the research methodologies for the primary research. Research Methodology Before research is undertaken it is important to determine the general approach that is to be used.
Looking at the goals that are to be attended and the way research can help attain these goals. Research Framework Eriksson and Wiedersheim-Paul (1997) has stated that the purpose of the research is to tell the readers of the paper the intentions of the study and how these results can be utilised. With the understanding of the purpose the methodological design can take place. There are three general structures that can be used in research terms; exploratory, descriptive and explanatory (Eriksson and Wiedersheim-Paul, 1997).
Where there is a problem and the purpose of the study is to increase the understanding of that problem the exploratory approach can be very useful (Saunders et al, 2000). This is an approach that looks at gathering the basic data regarding the problem and the scenario and has the aim of developing that knowledge and proposition that can then be used for further research (Yin, 1994). The descriptive study approach looks at the isolated empirical generalisations that have already been study and published and tries to explain these (Miles and Huberman, 1994).
This can be defined as making “complicated things understandable by reducing them to their component parts” (Miles and Huberman, 1994). This is most suitable where the research is not to look at the research between the causes and the symptoms, but wants to understand, predict or decide (Eriksson and Wiedersheim-Paul, 1997). Explanatory research takes complex issues and then tries to make them more understandable by examining the connections between the different parts by examining cause and relationship (Miles and Huberman, 1994).
This is not looking for only traditional causal explanations but seeking to determine them in the way that that research question has been phrased and how the specific factors impact and interact with each other. When this approach is utilised the research need to consider the use of more than a single research model as this would not be sufficient to explain the observed phenomena and create a full explanation that can be tested.
It is worth noting that when undertaking research there is no need for the methodology to be confined to only a single approach, indeed it has been suggested that mixed approaches can be very valuable with the research itself broken down into three stages of exploratory, descriptive and explanatory (Reynolds, 1971). The stage of the research depends on the existing knowledge within the area, at the explanatory stage the aim is for the researcher to obtain a better understanding of the area of research.
In the following stage there is the aim to develop descriptions of the patterns which are observed that may have been indicated as expected at the exploratory stage (Reynolds, 1971). The last stage is that of explanatory at which stage the research that has already taken place can be utilised in order to determine the relationship between the cause and effect this can then provide a researcher with a cycle of construction of the theory, hypothesis testing and the restatement of the theory (Reynolds, 1971).
This research has the aim of collecting data, assessing and describing the patterns seen in the Jamaica banking provisions for its Employees and how the issue of job satisfaction is important. When considering the way in which the research can be undertaken there are two philosophies; inductive and deductive, but both are able to produce valid results (Eriksson and Wiedersheim-Paul, 1997). Where there is the use of empirical data and research will draw conclusions from this empirical data and then applying and apply theories and models observed in the real world this is an inductive approach.
Where there is an approach that makes use of existing theories and testing these theories this is deductive research (Eriksson and Wiedersheim-Paul, 1997). This is a part of study where there is the publishing of data and results of different countries are available, but there is not a great deal of empirical evidence to test, as such the approach will be inductive with the use of empirical data and the applications of observed theories to try and explain the differences seen in the student performance levels.
This approach to the research will ensure that the attention is focused on what can viable be achieved within the time frame and budget and also on the context of work that has already taken place, which is lacking. The data has to be gathered from the population being sampled, so the next consideration is the way in which sampling will take place. The decisions to use questionnaire is justified as this is the most effective way of ensuring that there is the targeting of the correct respondents and there is also control over the performance of the surveys.
The next section will look at why we are using questioners as the research tool Research Tool Questionnaires are one of the most popular ways in which data may be collected. These are used for a range of purposes and will often be the tool used to collect quantitative data. But it is worth mentioning that although this may be classified as mostly a quantitative process of collecting data it may also be a qualitative method, depending on the number, depth and style of the questions.
Quantitative research over here has relied more on the large amount of results, with the recorded data being mainly numerical in type and has been suitably used as a method of determining cause and effect relationships (Curwin and Slater, 2001). But this qualitative research is narrower and more concentrated by looking to information to be gained in-depth from a smaller quantity of subjects (Curwin and Slater, 2001). Since the data over here require phenomenal research of Grounded theory as well as historical research hence this Qualitative research is best suited for getting applied (Curwin and Slater, 2001).
Since the research undertaken over here has been basically done on analysing how the Employee growth has effected change in the behaviour pattern of Jamaican Banking industry as well as the change in the social pattern of the state, the study of the same is going add something to the existing knowledge. Therefore, we must consider the different aspects of research that will characterise a research project as valid; these may include issues such as reliability, validity, objectivity, causality and bias.
These may be seen in the questions that are asked, the way they are asked through phrasing and order, and in questions that are not asked. Therefore the questionnaire has been designed while considering the full possibility of getting certain answers, but in what type of answers are attained and to be seen as impartial. There are different aspects that need to be considered so that it will be accepted and good solid research. The formulation of a questionnaire may be inherently biased and reflect the desired out come rather than an objective research method.
Reliability may be perceived as one of the easier objectives. This is often seen as the ability of the same study to be conducted and the results to prove the same, however repeatability does it mean that a research project is valid and can be relied upon (Denzin, 1978). We have considered other aspects of the research rather that just reliability in order to make the project valid. Reliability must also come from the level of confidence that may be seen in a study from the objectivity or recognised subjectivity which may be seen to arise.
The project have been undertaken from a subjective perspective and those asking the questions if a face to face method is used, do not known the study purpose. The first stage to recognise that subjectivity is present and then recognised in order to overcome. If we set out to prove a model is true it is easy for the research to be biased, just as if we are looking for specific information other aspects may be ignored or forgotten. As such although the researcher may be biased there are certain measures that can be taken to ensure the research model used can be seen to have been implemented objectively.
In designing the questionnaire it is likely that there will be others involved either by way of observation or the correlation of the data. When designing the research questions, there should be no indication of the final purpose of the study, in this way we might limit the bias of the respondents and ensure there is a lower risk of the cross contamination of bias (Denzin, 1978). However, it must also be recognised that we can never completely extinguish research bias. With the consideration and design of the research tool the next consideration is that of sampling.
Sampling When looking at the way a sample should be chosen there will be the desire to choose a sample that is representative of the population that is going to be targeted. If the population is not representative then the results will not be representative. The way in which we need to consider the population is to look at the way sampling will be used. In this scale we need to talk to people who are working as a professional in one of the nineteen banks in Jamaica. The first aspect is whether probability or non-probability sampling should be used.
Probability sampling is a more random sampling style, the basis of this is that the selection of each respondent is a matter of chance and that all respondents will have an equal chance of being chosen, in many cases this can be a known probably of the respondent being selected. There are a number of sampling techniques that fall into this category. These include simple random sampling, systematic sampling, stratified and cluster sampling. In total there are more than 30 probability sampling techniques (Malhotra, 1999).
Random sampling is best defined as a sampling method was any member of the population will not have a zero chance of selection where the chances can be calculated. Looking at these different methods, simple random sampling is the most basic. This is where all members shave the same chance of being chosen and where the knowledge of another population member being chosen for sampling will not alter the odds. A method of undertaking this is choosing names out of a hat or any other random method.
Stratified sampling is where the population of the sample is separated into distinct groups which do not overlap; this is where stratification is seen. There will then be samples taken from each of these groups. For example, this may be a sample that is divided into male and females, likewise we may look at those who can and do use the internet and those who do not. There may also be the use of age groups, they key is no overlaps between the groups. The next potential method may be multistage sampling.
This may consist of two, three or more stages in the sampling. The process will start with the division of the population into non-overlapping groups as seen with the stratified sampling. However this is only the first stage, following this a sample of the first stage units will be chosen according to different criteria. This can be complex, but we could look at using this to use the second stage to ensure that we can gather results from those who are relevant to the study, such as Employees at the managerial level and have a list of employees working under him.
The simplicity of these methods is attractive and the use of random sampling with the use of strafed sampling may be attractive, but there are also difficulties as we are reliant on the probably that the sample will be representative of the sample population rather than an assurance existing that they do not appear to be skewed. In a small sample population it is more important to seek to eliminate this bias than in larger sample groups.
In the general population there may need to be a large number of survey’s undertaken before there is the random finding of a respondent that is a ban employee. In order to overcome this danger we will look at non probability sampling techniques as these will have a level of intervention to ensure the sample is representative. The most common is quota sampling, but this will also include judgement sampling, convenience sampling, random route sampling and snowball sampling (Talmage 1988). Quota sampling is exactly as it appears to sound, there are the definition of quotas to be used.
In many cases these will consist of gender, age groups and social class quotas, which may or may not have crossovers. This is used widely in commercial settings but this is also a method that is not supported as widely in academic circles (Ojeda and Sahai, 2002). Opportunity sampling is where an opportunity is sought to interview or gain the opinions of a particular group and will then place themselves or the survey in the position of those who may be most appropriate for that study (Talmage 1988).
For example, if the employee a bank is placing the interviewer near the business section of the bank may be advantageous. Random route sampling where is there is a route planned and the interviewer may travel that route to gather data and hopefully meet potential subjects. This is a very random method of sampling and is also difficult to gain the correct numbers where there is quantitative data required. Snowball sampling is a method that can be used when there is a scenario of a large level of data being required in a short period of time (Talmage 1988).
This may be used with a few key people being approached and interviewed which will then lead to more potential candidates being identified (Talmage 1988). Self section is also a non random type of sampling as the respondents will volunteer themselves and a last type we may consider is that of plausibility sampling which is defined by Talmage (1988) as “a sample selected because it appears plausible that the members are representative of a wider population, without any real evidence” (page 82).
With each of these methods there are advantages and disadvantages, we need to consider which is going to be most appropriate considering the population required and the sample and the sampling method. Looking at all of these it appears that the most appropriate method will be that snowballing, it is not the fastest but it is likely to provide the best results in identifying potential respondents for the research.