2. METHODOLOGY
2.1 Research Design
In terms of this case, descriptive research design, which is part of conclusive research design, would be conducted to evaluate the impacts of the Women in the Smart State Directions Statement 2003-2008 on the Queensland women. This research design method is based on large, representative samples and the data gained are subjected to quantitative analysis; in addition, the findings from this research are considered to be decisive in nature since they are used as input into managerial decision making (Malhotra et al, p105). In particular, this research design is used to be able to integrate findings from different sources in a consistent manner as well as estimate the percentage in a specified population showing a certain form of behavior (Malhotra et al, p105). With regard to the case, the campaign has been implemented for a couple of years and the target population, Queensland women, might have different opinions on it. Therefore, via the quantitative method the government may collect large amounts of information from enough members of the target population so that inductive logic and probabilistic inferences can be understood.
2.2 Method
Descriptive research designs have come to be viewed as the different survey research methods for collecting quantitative primary data from large groups of people via the question/answer protocol process (Bush et al, 2003, p.255). The data collection procedure emphasizes that the respondents will be asked sets of standardized, structured questions about what they feel, think and do. When it comes to this case, compute-assisted telephone interviewing (CATI), one of quantitative survey methods, would be adapted to collect quantitative data. The advantages of this data collection method are that the interviewing time is reduced, data quality is enhanced, and the difficult steps in the data-collection process, coding questionnaires and entering the data into the computer, are reduced (Malhotra et al, 2006, p.254).
3. SAMPLING
3.1 Sampling design process
The sampling design process consists of five steps which are: (1) Define the target population; (2) Determine the sampling frame; (3) Select sampling technique; (4) Determine the sample size; and (5) Execute the sampling process.
3.1.1 Define the target population
The target population contains four parts that has to be mentioned. According to this case, the five goals of Queensland Government’s campaign are to improve the awkward situation of women such as under-representation, inequity and need. In addition, the Direction Statement involves three different age groups, young women, women and aged women. In brief, the target population should be qualifies as follow:
- Queensland women aged over 14.
- Queensland women who have been living in Queensland at least 4 years.
Moreover, the sampling unit is consider to be the household telephone numbers in that the interview would be conducted by the telephone. Also, the survey will be implemented only in Queensland for two months. For this reason, the extent is Queensland and the time is one month.
3.1.2 Determine the sampling frame
Sampling frame: Computer program for randomly and efficiently generating telephone number.
3.1.3 Select sampling technique
In terms of sampling technique non-probability sampling is not suitable to use in that non-probability samples may produce good estimates of the population characteristics. Nevertheless, they do not allow for objective evaluation of the precision of sample result. Furthermore, since there is no way of determining the probability of choosing any particular element for inclusion in the sample, the estimates obtained are not statistically projected to the population (Malhotra et al, 2006, p.367). In comparison with non-probability sampling, probability sampling, sampling units are selected by chance. Moreover, it is possible to pre-specify every potential sample of a given size which could be drawn from the population, and the probability of choosing each sample. For this reason, the researcher could make inferences or projections about the target population from which the sample was drawn (Malhotra et al, 2006, p.367).
Proportionate stratified sampling with two-step process, one of probability techniques, will be conducted for this case. The population is divided into subpopulations or strata, and every population element should be assigned to only one stratum no population elements should be neglected. Next, elements are chosen from each stratum by a random procedure, and SRS will be adopted to implement (Malhotra et al, 2006, p.374). In addition, the elements within a stratum should be as homogeneous as possible; however, the elements in different strata should be as heterogeneous as possible (Malhotra et al, 2006, p.374).
In practice, the target population will be divided into three stratums, which are women aged 14 to 17 (including 14 years old), women aged 18 to 59 and women aged over 60. The reason to divide the target population into three age groups is that the government has been launching various campaigns for them. Therefore, most of them could answer the questions. According to the projected population by age, Queensland, 2006 to 2051, in Queensland the ratio of women aged 14 to 17 to women over 60 is about 1:2, and the ratio of women aged 18 to 59 to women over 60 is around 7:2. The table below provides the information to easily understand the sampling technique, proportionate stratified sampling, in terms of this case.
3.1.4 Determine the sample size
Malhotra et al (2006) mentions that national studies of households typically have samples ranging from 1000 to 2500 or more, while regional studies have 200 to 1000 or more samples (p.366). Therefore, the sample size for this case is set to be 1000. The table below shows the number of sample that will be drawn:
3.1.5 Execute the sampling process
Execution: Use a computer program to randomly generate a list of household telephone numbers and conduct the interviews using a computer-assisted telephone interviewing (CATI) system.
3.2 Data collection methods
3.2.1 Instrument of data collection method
Computer-assisted telephone interviewing (CATI) will be utilized as the data collection method. The computer automatically dials the prospective respondent’s phone number and provides the interviewer with the appropriate introduction and computerized questionnaire on the screen. Therefore, the interviewer reads questions for the respondent and simultaneously records the respondent’s answers into the computer by use of a keyboard. It seems that CATI system could save time and decrease the mistakes.
In addition to data collection method, questionnaires will be used as the instrument for this case. The design of the questionnaire is based on some variables, and the Likert scale will be used for this questionnaire with five response categories which range from ‘strongly disagree’ to ‘strongly agree’ (Malhotra et al, 2006, p.333). Also, there are several basic information questions for the respondents to answer in that the basic information is important to generate variables. Many independent variables, such as age, income, family and so on, would influence the dependent variable (attitude) and relationships between these variables are research object. The question will be divided into some sections in the questionnaire, and each section stands for one independent variable. As a result, via CATI the data could be collected easily and reveal the thinking of respondents about the particular independent variable.
3.2.2 Data gathering process
Before conducting the procedure of gathering the data, the selection, training, supervising, validating and evaluating of interviewers would take one week to accomplish. Then, five interviewers will be employed for gathering data via CATI, and the data gathering period will continue to implement for nearly two weeks. Due to the use of Computer-assisted telephone interviewing (CATI), the interviewers can record the respondents’ answers directly into the computer while they are asking the respondents the questions. Therefore, the data will be collected with fewer mistakes.
3.3 Data analysis methods
After collecting the data the researchers can make use of various data analysis techniques to produce useful information for decision maker to operate a business. With regard to this case, the data analysis technique, such as ANOVA, would be adopted and it is used to identify differences between women. It seems that the information produced by this data analysis technique is useful to consider whether the state government should continue the Directions Statement or not.
4. DATA ANALYSIS
ANOVA is a statistic technique used to analyze data and examining the differences among means for two or more population. (Malhotra et al, 2006, p.590). In its simplest form, ANOVA has a dependent variable that is metric (measured using an interval or ratio scale), while there must be one or more independent variables which is all categorical, non-metric (Malhotra et al, 2006, p.590). The outcome of ANOVA provides a more detailed analysis or information for the decision maker. That is, ANOVA might increase the comprehensive and completeness while the management is making the decision.
This proposal is to evaluate the impact of Directions Statement, and there are lots of possible factors that would influence the women’s attitudes. Therefore, these factors should be examined to produce more information for the government via ANOVA. For example, the government launched an anti-smoking campaign and promoted it for a period of time; however, the attitude toward the implementation of this campaign might alter in different age groups. For that reason, the state government could understand that the attitude might be influenced by the age groups, and then it might improve the campaign to fit for most of women’s expectation.
5. LIMITATIONS
In terms of the marketing research design, survey method has the ability to accommodate large sample size and enhance the researcher’s ability to make probabilistic inferences about the defined target population as whole. However, it still contains some limitations while the researchers are conducting it. The limitation would be like that developing perfect survey instruments, such as questionnaire design, is considered to be difficult. When it comes to the case, the questionnaire is designed for the respondents to answer; on the other hand, not all of them can really comprehend what the nature of question is. Therefore, they may ignore the question or reply it at random. Moreover, in comparison with exploratory research, it is not simple to understand what the respondents think in their mind. For example, the respondent just can answer the questions within the provided choices. As a result, they might select a choice that is opposite of what they want.
When it comes to this case, the interviewers would make use of the computer-assisted telephone interviewing (CATI) in order to accomplish the questionnaire. Nevertheless, for the researchers CATI systems have two limitations. At first, the investment in computer, software to control hardware, monitor calls and record responses in real-time fashion is still high and speedily changing. Secondly, the interviewers need to have particular computer skills to run this type of survey.
PART B
T-Test Output
1. The data analysis identifies that on the question of “Assessment tasks” there are significant differences between Full Time and Part Time Students (sig. 0.007).
2. The data analysis identifies that on the question of “Feedback from students used” there are significant differences between Full Time and Part Time Students (sig. 0.036).
3. The data analysis identifies that on the question of “Overall importance of unit” there are significant differences between Full Time and Part Time Students (sig. 0.038).
4. The data analysis identifies that on the question of “Teaching staff-developing knowledge, understanding and skills” there are significant differences between Full Time and Part Time Students (sig. 0.017).
5. The data analysis identifies that on the question of “Overall importance of teaching” there are significant differences between Full Time and Part Time Students (sig. 0.020).
ANOVA Output
The data analysis identifies that on the question of “Teaching staff understanding expectations” there is a significant difference between students who commence their studies in 2001 and all other years (sig. 0.000).
References
Hall, J., Malhotra, N., Oppenheim, P., & Shaw, M. (2006). Marketing Research 3rd. Australia: Pearson Education Australia.
Hair, J., Bush, R., & Ortinau, D. (2003) Marketing research: within a changing information environment 2nd. London: McGraw-Hill/Irwin
The State of Queensland (Queensland Treasury) (2007). Office of Economic and Statistical Research: Results of the 2004 Queensland Survey of Gaming Machine Venues. Retrieved May 17, 2007, from
http://www.oesr.qld.gov.au/queensland-by-theme/demography/population/tables/erp/erp-sd-ssd-qld/index.shtml