Quantitative Methods for Computing - Case Study Example

Quantitative Data Collection Methods Quantitative research design involves the use of numerical data to come up with answers to specific questions. In this design technique, descriptive statistics (mean, variance, standard deviation, range, kurtosis and skewness) describe a group of respondents (Wholey, Hatry & Newcomer 2004). To make this a reality, inferential statistics, statistics results based on a sample representing the whole population are used. Depending on the kind of population one is dealing with, a sampling technique is selected based on the traits of the population (simple random sampling if the population is homogenous and stratified sampling if the population is heterogeneous). The results of a certain sample can be generalized based on the method of collection and whether the right methodology was followed. Quantitative data is data that can be counted or which can be expressed numerically. Quantitative data is collected using experiments and is in turn collated, cleaned, entered into a computer, manipulated and analyzed statistically for meaningful results. Quantitative data is normally presented in form of graphs, charts and tables since the three form a bird’s eye view of the data and are easy to comprehend even by non-statisticians. Quantitative data is mostly used since it is specific, can be replicated over-and-over and it is easier to generalize quantitative findings than qualitative findings. According to Freed, M. et al. (1991), data collection is the most important part in any research study since the accuracy of the collected data has positive impacts on the results of the study. Inaccuracies in data collection lead to invalid results which when generalized at times leads to poor judgment and in-valid generalizations. Data collection methods differ depending on the objectives of the survey, bound on the error of estimation and type of analysis to be done. Quantitative data is normally meant for testing different formulated hypotheses. As Freed put it, the data collection method depends on the researcher’s question and that study participants are randomly assigned the different survey conditions to ensure that the element of biasness is minimized as possible. Further, to him, the researcher collect data controlling for the impact of the research dependent variables (DV). In quantitative research design, the willingness to generalize the results of the sample to be that of the whole population, then a larger sample is collected as this minimizes the error of the various descriptive statistics estimations. In such a case, the researcher employs the use probability sampling (sampling based on the chance of being selected) to select participants. Quantitative data collection techniques which can be applied to collect data for the “Human Dependence on Computers study” include recording experimental readings, observations (e.g. counting number of participants to a certain computer-use event at different time frames and comparing the different readings), secondary data from different systems, in depth questionnaires (closed-ended, open-ended, face to face questionnaires), interviews (face-to-face, online and telephone) (Gore & Altman 1992). Quantitative data from experimental recordings are normally recoded in checklists from time to time and are then compared over time. A checklist may in a manner that the level of computer use has different grades and the user is graded according to his/her abilities. The use of observations is another technique which can be used to collect quantitative data for the research study and involves observing different users of the computer and gauging the level of usage. Normally, collecting data from experiments and observations is advantageous in that the data collected is first hand, tangible, reliable and valid. In-depth interviews are another method used in the collection of quantitative data. Face-face interviews enable the researcher to create a relationship with the interviewee and their use is advantageous since it allows the researcher to seek for clarifications and more information in case of leading questions. Telephone interviews on the other hand afford the interviewer less time and resource requirements. This is so because; the researcher is able to get all the necessary materials/information by calling the respondents from time-to-time for follow-up with very little funds. The use of questionnaires is another method of collecting quantitative data. Normally, questionnaires are used to collect data from so many respondents in one go and normally saves the researcher time and finances. Both closed-ended and open-ended questions are used and names or any other contact information are not included giving the respondent the freedom to respond to the questions even controversial ones due to the feeling that their responses are confidential. Open-ended questions are used in case the researcher would like different opinions of the different respondents. Such questions data are then collated and coded to conform to quantitative data requirements before analysis. However, closed-ended questions are mostly used in questionnaires as they restrict the responded to give certain answers and are easy to analyze and quantify. The only disadvantage with questionnaires is that some respondents may fail to return the questionnaires in case of the mailed ones thus affecting the response rate and in turn the sample precision. To get a deeper understanding of the “Human Dependence on Computers”, data was collected using a questionnaire with 10 questions; gender, marital status, age, level of education, professionalism, hours of computer use per week, whether during orientation after first appointment one was oriented on computers, whether one has been trained on computers since employment and whether thy consider computers as being effective in service delivery. The results follow below. Results Table 1: Gender Frequency Percent Male 24 44.4 Female 30 55.6 Total 54 100.0 In terms of gender representation, females were many than males (55.5% Vs 44.4%). A binomial test was conducted to determine whether the difference in representation was significant at 5% level of significance (see table 2). Table 2: Binomial Test Category N Observed Prop. Test Prop. Asymp. Sig. (2-tailed) Group 1 Male 24 .44 .50 .497(a) Group 2 Female 30 .56 Total 54 1.00 a Based on Z Approximation. From table 2 above, the representation of males and females is not significant at 5% level of significance [Test proportion = .5, p = 0.447]. Further, from the results, the use of computers is the same for males and females [Chi-square (2) = 0.710, p = 0.701] (see below). Table 3a: Chi-square Test Value Df Asymp. Sig. (2-sided) Pearson Chi-Square .710(a) 2 .701 Likelihood Ratio .708 2 .702 Linear-by-Linear Association .286 1 .592 N of Valid Cases 54 a 1 cells (16.7%) have expected count less than 5. The minimum expected count is 4.89. Table 4: Marital Status and How long do you use a computer per week? Cross-tabulation How long do you use a computer per week? Total 1-15 Hrs 31-45 Hrs 46-60 Hrs Marital Status Single 11 0 0 11 Married 0 26 12 38 Widow/Widower 0 4 1 5 Total 11 30 13 54 Table 3a: Chi-Square Tests Value Df Asymp. Sig. (2-sided) Pearson Chi-Square 54.353(a) 4 .000 Likelihood Ratio 54.894 4 .000 Linear-by-Linear Association 30.143 1 .000 N of Valid Cases 54 a 5 cells (55.6%) have expected count less than 5. The minimum expected count is 1.02. From the above tables (table 3 and table 3a), it is evident that 70.4% of the participants were married while 20.4% were single. 9.5% of them were widow/widower. Further, it is evident that there is a significant association between marital status and hours spend on computers with the married participants spending more time that single and widows/widowers [Chi-square = 54.353, p Read More