Research And Statistical Method For Business: Sample Size, Sampling Method, And Measures Of Variables

Sample Size

Discuss about the Research and Statistical Method for Business.

Businesses need research and statistical methods to collect data, store, analyze and interpret data to make informed management decisions. Research is typically conducted to investigate hypothesis and establish truths about events (Hair, Wolfinbarger, Money, Samouel, & Page, 2015). Many business models have been developed to inform research and statistical methods of data collection and analysis including the “Research Onion” which is used to describe the research method. The research onion was first created by Mark Saunders in 2007 and later developed as a tool for formulating the research methodology (Saunders, Lewis, & Thornhill, 2015).

Research in business is used to explain and analyze business events. The process of research begins with planning, data collection, analysis, discussion, interpretation, and recommendations for future reference. For optimal results, it is important to study a wide range of variables including people, factors and their interaction (Hair, Wolfinbarger, Money, Samouel, & Page, 2015). Appropriate research should be generalizable and replicable, and its output should exceed the input. This paper answers questions on business research methods with regarding a case study by De Witte, De Cuyper, Handaja, Sverke, Naswall, & Hellgren (2010). De Witte and his team sought to understand the association between employee’s perception of quantitative and qualitative job insecurity with job satisfaction, and psychological distress in a cross-sectional survey of the Belgian banking sector.     

Sample size refers to the representative sample of the general population (Spradley, 2016). Typically, a sample size of between 30 and 500 is acceptable but larger sample sizes give more data for examination and analysis (Bryman & Bell, 2015). In this case study, the sample size is 15,000 representing 21% of the total population which is 69,000. This sample size is valid and reliable as it will allow the researcher to make conclusions regarding the business phenomenon (Sekaram & Bougie, 2016).  As a result, a sample size needs to be large enough to allow for the estimation of closely related variables with minimal variance. However, the sample size should not be larger than the general population no matter what. The sample size usually varies according to the population size, confidence level, research objectives and other externalities such as the demographic composition of the population. 

Still, the researcher should avoid using a sample size that is too large or too small as it may mislead the study. For instance, the research may accept and hypothesis when in fact they should be rejecting it (Sekaram & Bougie, 2016). Massive sample sizes may reach the significance levels misinforming the research making it less trustworthy. For instance, the researcher may interpret a relationship between variables in a manner parallel to the actual situation.    As such, no sample that is too small or large that is helpful to the research study. Precision should be used to guide the researcher in achieving an efficient sample size.

Sampling Method

According to Sekaram and Bougie (2016) suggest that the sample size should not be less than 30 or more 500 for it to be acceptable is research. Kothari, on the other hand, asserts that the right sample size is 30 % of the general population. Further, in multivariate analysis, the sample size should be larger than the number of variable in multiple times. However, in simple studies under controlled environments can achieve success with smaller sample sizes of between 15 to20 subjects.  

The case study of importance to our case study is the issue of heterogeneity of the population. Samples of heterogeneous data were collected from the highly different samples are highly diverse as compared to homogenous samples; therefore, “the greater the heterogeneity of a population, the larger the sample will need to be” Bryman and Bell (2015).

Considering the factors discussed thus far, the sample size of our case study (15000) was necessary. The study sampled 63 banks across Belgium which were mostly heterogeneous were largely heterogeneous and demanded of a more significant amount from the researcher. The research has many variables requiring a large sample size according to the rule of thumb a large sample size increases precision and is a more reliable and trustworthy process.

De Witte et al. (2010) used simple random sampling to select it to sample size to form the study population. Simple random sampling has no strata, and it allows every member of the community an equal chance at being selected. Simple random sampling is a statistical method that gives each subject in a variable a fair opportunity of selection (Bryman & Bell, 2015). For example, in simple random sampling names written on several pieces of paper with similar characteristics when put in a bowl and stirred thoroughly and one piece is selected as a representative. The research needs to mix the contents of the container each time a representative of the population is to be selected. Stirring ensures that even the subjects at the bottom of the bowl get an equal chance of being picked. Simple random sampling is commonly used in raffle ticket competitions.

According to Zikmund, Babin, Carr, and Griffin (2013), the simple random sampling method is the most straightforward sampling technique because there is only one stage of individual representative selection. Samples selected through random sampling are representatives of the general population and can be used to determine the situation within the population. A disadvantage of the method is that it may not be the best because of the time and resources used to repeat each selection procedure, especially where the sample size is as big, and the subjects are dispersed like in the case study.

Measures of Variables

Another advantage is the ability of simple random sampling to eliminate human bias in sample selection. The method does not allow the researcher to select his/her sample based on personal merit such as being attractive, among other (Sekaram & Bougie, 2016). Further, the subjects of study are selected without their knowledge and can not influence the selection process just like in job applications where only a few people from a list of numerous applicants get selected for an interview. They only get to know about their selection when they are informed to attend the interview. Seemingly, the advantages outweigh the disadvantages of the sampling method.

Variables are measured using tools either developed by the researcher or using models and concepts developed and tested for validity and reliability by other researchers.  Tools used in measuring variables should be scrutinized thoroughly to avoid mishaps (Bryman & Bell, 2015). However, this is not always the case as some researchers hurriedly adopt measures of variables without judging their applicability efficiently. They simply assert that the measures are appropriate by judging their straightforwardness and ignore the tests on validity and reliability.

In the study, multiple measures have been used to derive answers to the research questions.  Bryman and Bell assert that when multiple measures are used,  face validity and internal reliability should be examined (2015). In reality, this is not always the case as measures of variables are at times not tested to establish the internal reliability. Reliability and validity precede each other, and when one measure is unreliable, it automatically becomes invalid. Measure that is unreliable and invalid produce varied results that fluctuate every often due to the differences contained during different times. If a measure is invalid, the subjective observations appear to be inconsistent since researchers disagree about their observations making the measure unreliable.

The case study uses multiple measures of variables to understand the qualitative and quantitative job insecurity, psychological distress, job satisfaction and several control variables on social demographics and work-related factors. The study does not indicate whether the reliability and validity of the measures of variables used were tested for consistency and effectiveness. However, Saunders et al. suggest that questionnaire, interviews, and observations are appropriate tools for measuring variables. Further, the reliability and validity of measures depend on the ability of the instrument to provide answers to meet the research objectives.

Usually, in survey designs, like in the study, the measures are tested for reliability and validity through pilot studies conducted before the actual survey to eliminate ambiguities and other factors that may cause conflict and confusion between the researcher and the respondents (Hair et al., 2015).  Fortunately, come of the measure used in the study had already been using in previous research and could be readily adapted with trust to produce data that is trustworthy. For instance, De Witte had already developed and used the four items measure in 2010.  The standards, however,  need to be tested to ascertain their level of reliability and validity.   

Data on social demographics is vital to the researcher as it helps in deciding who to include and who to omit from the research.  Demographic variables give the biodata of the individuals in the sample and include personal information such as name, sex, education level, income level, status, occupation, and age. The demographic tells the population’s characteristics allowing researchers to assess their subjects for inclusions of data and survey delineation into groups (Bryman & Bell, 2015).

The topic of the study determines who should be included in the survey. For instance, in the case study, the researcher might decide to categorize the information based on age demographics, experience, and occupational status. Still, the researcher might decide to use the demographic data on the level of education to group the respondents for ease of analysis. Data on social demographics direct eh researcher to determine who to include in the survey.

The demographic information collected after the return of interview results are used to divide the data into groups for ease of interpretation (Saunders, Lewis, & Thornhill, 2015). The researcher in the case study can cluster the individual responses in the survey using sex, as either female or male. De Witte et al. can also use the demographic data to compare and contrast data within the population sample. Although the researcher may want to understand his or her population more, they should not give in to the temptation to ask more questions that are necessary because the participant may feel that their privacy is infiltrated. Of course, the study does not want to aggravate the respondents as this may affect their output ultimately reflecting in the information given in surveys. The researchers should decide the demographic question to be included in the survey to avoid invading the respondents’ privacy and degrading their level of confidence in the research. Ethical concerns in research require that the respondents be assured of privacy and confidentiality of their information. The participants may be antagonistic when they perceive their privacy to be at risk.     

Research design refers to the process undertaken during the research. The design is a guiding framework that describes how the researcher decided to use specific approaches to others and is a function of the research objectives (Bryman & Bell, 2015).  According to Saunders et al. there are several research designs including explanatory, exploratory, descriptive and inductive designs (2015). The case study employed a descriptive survey research design across the Belgian banking sector.

Descriptive design describes the experience of the study respondents and closely relates to ethnographic studies except that the descriptive design facilitates a quantitative framework (Bryman & Bell, 2015; Saunders, Lewis, & Thornhill, 2015). The explanatory model explains the demographics of the participants efficiently to determine how variables influence each other. On the other hand, the exploratory design functions to explore the available information regarding the research topic and inform the literature review part of the study.

Conclusion

This paper has adequately answered the questions regarding the methodology used in the research covering the sections of sampling and sample size, measures of variables, a collection of data on social demographics and the research design of the study. The methods to be used in a research study are determined based on the topic and study objectives. The researcher should first identify the variables to be tested and the population sample.  The study should be generalizable and replicable. It should also observe ethical concerns in social research.

References

Bryman, A., & Bell, E. (2015). Business research methods. USA: Oxford University Press.

De Witte, H., De Cuyper, N., Handaja, Y., Sverke, M., Naswall, K., & Hellgren, J. (2010). Associations between quantitative and qualitative job insecurity and well-being: A test in Belgian banks. International Studies of Management & Organization, 40(1), 40-56.

Hair, J. F., Wolfinbarger, M., Money, A. H., Samouel, P., & Page, M. J. (2015). Essentials of Business Research Methods (2nd ed.). New York: Routledge.

Saunders, M., Lewis, P., & Thornhill, A. (2015). Research Methods for Students (7 ed.). Pearson.

Sekaram, U., & Bougie, R. (2016). Research methods for business: A skill building approach. United Kingdom: JOhn Wiley & Sons Ltd.

Spradley, J. P. (2016). The ethnographic interview. Waveland Press.

Zikmund, W. G., Babin, B. J., Carr, J. C., & Griffin, M. (2013). Business Research Methods. Mason, OH: South-Western Cengage Learning.