The operationalization of variables is a fundamental step in any research process. This step entails the development of procedures that will result in the required measurements of dependent and independent variables. Extraneous and moderating variables must also be operationalized. According to Babbie & Mouton (2001), the operationalization of variables generally entails the construction of definite, tangible measurement procedures that will inarguably represent the concepts of interest.
The sample study has three dependent variables – turnover, morale, and job satisfaction – and several independent variables. According to the hypothesis and research questions, service satisfaction and the desire to quit service forms the dependent variables, while payment levels and tenure conditions form the independent variables.
The above variables can be measured by nominal, ordinal, interval, and ratio measures. Nominal measures are good at reflecting the categories of soldiers in relation to marital status – married or single. Ordinal measures will greatly assist in operationalizing variables such as service satisfaction and tenure conditions as these particular measures are able to logically rank the desired attributes. For instance, you can use an ordinal measure of ‘very satisfied, fairly satisfied, neither satisfied nor dissatisfied, fairly dissatisfied, and very dissatisfied’ to rank the service satisfaction levels and tenure conditions.
Interval measures can be effective in operationalizing variables such as payment levels and desire to quit service as they attach a meaning on the intervals between various attributes of interest. For instance, a soldier who receives less pay package as reflected on the interval measure is more likely to leave service than one who receives more pay. Lastly, ratio measures can be used to quantitatively study disparities between attributes as it bears a true zero point (Babbie & Mouton, 2001).
In this study, demographic variables such as age, marital status, gender, level of education, family size if married, base locations and current pay package must be measured as they directly relate to the variables under study. These demographic variables are fundamental in segmenting the reasons that soldiers consider most when making important decisions such as quitting employment, payment considerations, and service satisfaction (Kaliyamoorthy & Prem, 2007).
For instance, a soldier will tie his pay requirements to his or her educational status, marital status, and size of the family. In respect of this, demographic characteristics form a critical component of any research process. On most occasions, demographic data are able to reveal issues of interest such as important opinion dissimilarities between men and women regarding variables such as service satisfaction and desire to quit service.
Attitudes and opinions must be sought from the respondents if variables such as service satisfaction, tenure conditions, and payment levels are to be successfully measured. Items such as ‘very satisfied, fairly satisfied, none, fairly dissatisfied, and very dissatisfied’ can be used to capture critical information on payment levels and service satisfaction (Babbie & Mouton, 2001). Neatly organized items bearing a common response set should be used to seek responses to each of the above variables.
The Guttmann scale, developed in the 1940s, can be used to evaluate if any relationship exists within a cluster of items used to measure variables such as the desire to leave employment. In most cases, the items are ordered in an ascending manner (Page-Bucci, 2003). For instance, soldiers may be asked to indicate what they think about promotions against items such as: ‘promotions are not given on merit; promotions are usually delayed; promotions do not necessarily change work environment; promotions make one achieve long-held dreams, and promotions are purely based on merit.’
To make it easier, values should be accorded to each item on the measurement scale. For instance, in measuring job and tenure conditions, the soldiers can be asked to rank themselves against a categorized scale that is capable of measuring their attitudes and opinions towards the job. They can be asked to state what they look for in the job, giving the following categories:
- well-paying job;
- work that offers feelings of accomplishment;
- work that has minimal supervision;
- work that is pleasant;
- work that is steady.
A Likert scale consisting of 1= strongly agree; 2= agree; 3= neither; 4= disagree, and 5= strongly disagree, can be effectively used to measure attitudes and opinions about tenure conditions and desire to quit service (Page-Bucci, 2003).
Variables such as payment levels and turnover rates may require organizational data or archived data since the respondents may not be in a position to offer fulfilling responses. Performance appraisal techniques and training methodologies may also require organizational data. The ability to integrate archived and organizational data with the mainstream data achieved from fieldwork is critical to the success of any research process (Moulton, Madnick, & Siegel, 2001).
This study intends to use a convenience sampling technique in the data collection process. According to Forzano (2008), this non-probability statistical technique concerns itself with the selection of specific observations aimed at yielding some detailed knowledge about a population of concern. Convenience sampling is justifiable since the study is dealing with a homogenous group – the army. In the same vein, taking random samples of the army would be almost impossible judging by its huge size.
In such circumstances, convenience sampling will enable the researcher to utilize the available respondents provided they discreetly fall under the classification of army personnel (“Convenience Sampling,” 2009). Judging by the nature of work that the US army is engaged in, it is fundamental to choose a method that will allow data collection based on the easy availability of the respondents.
Reference List
Babbie, V., & Mouton, C. (2001). Research design – operationalization of variables. Web.
Convenience Sampling. (2009). Web.
Forzano, G. (2008). Research methods for the behavioural sciences. Cengage Learning EMEA. Web.
Kaliyamoorthy, S., & Prem, K.S. (2007). “Influence of demographic variables on marketing strategies in the competitive scenario.” The Icfai Journal of Marketing Management, Vol. 6, No. 3, pp. 53-62. Web.
Moulton, A., Madnick, S.E., Siegel, M.D. (2001). Cross organizational data quality and semantic integrity: learning and reasoning about data semantics with context interchange mediation. Web.
Page-Bucci, H. (2003). The value of Likert scales in measuring attitudes of online learners. Web.