Discusses The Methodology Of Research Psychology Essay

CHAPTER 3RESEARCH METHODS
3.1 Introduction

Research is conducted to produce information in order to reduce the uncertainty and also as a tool to provide guidelines and helps in decision making. According to Zikmund (2003), research design consists of three types, which are exploratory research, descriptive research and causal research.

3.2 Research Design

The nature of the problem in a business research will help to determine the type of the research. The types of research includes exploratory, descriptive and causal. Exploratory research is the initial research conducted to clarify and define the nature of the problem. (Zikmund, 2003). Descriptive design is to describe the characteristics of a population or a phenomenon. (Zikmund, 2003).

Causal research is conducted to identify the cause and effect relationships among the variables when the research problem has been narrowly defined. (Zikmund, 2003). The research design of this research study is based on the causal research.

3.3 Sampling design
3.3.1 Target population, sample frame and sample size

In this research, the researcher focuses on the population of employees in the restaurant industry. From the population, the researcher provides a sample frame which is a target population. A sample frame is a comprehensive list of the elements. Therefore, the researcher gets the list of employees from the manager.

In order to get the sample size, the researchers had distributed one hundred and fifty (150) sets of questionnaires to the employees of the restaurant industry in Dataran Sunway, Kota Damansara. The sample size can help to determine either large or small population. In this research, the researcher focuses on the small population of the restaurant employees.

.3.3.2 Sampling Method

The sampling method can be classified into probability sampling and non probability sampling. Probability sampling is a sampling technique in which every member of the population has a known, nonzero probability of selection. (Zikmund, 2003). Whereas, the unit of samples selected in non probability sampling are based on the basis of personal judgment or convenience. (Zikmund, 2003). In order to have a random selection method, the researcher must set up some process or procedure that assures different units in researcher’s population have equal probabilities of being chosen. (Trochim, 2007).

In this research study, the probability sampling method is to be implemented to conduct the sampling process. The method used is the simple random sampling. Simple random sampling is a sampling procedure that assures each element in the population an equal chance of being included in the sample. (Zikmund, 2003). Simple random sampling is used because in this method there is only minimal advance knowledge of population needed. It is also easy to analyze data and compute error. Every employee in restaurant has equal chance of being selected which can increase the accuracy, relevancy and the credibility of the research.

3.4 Data Collection Methods
3.4.1 Primary data

According to Zikmund (2003), primary data are data gathered and assembled specifically for the research project at hand. There are numerous methods of collecting primary data, such as questionnaires, interviews, observation, case studies, diaries, and etc. Questionnaires method has chosen in order to gather information from the targeted respondents in this research study. This is because questionnaires method can cover a large number of people or organizations, and are relatively cheap as compared to personal interviews and telephone surveys and the respondents can provide feedbacks freely as there is no interviewer bias.

Information is collected by distributing the questionnaires to the employees in the restaurant industry. One hundred and fifty (150) sets of questionnaires have been distributed to the targeted respondents. The questionnaires consists of introduction part, followed by six (6) questions in Section A, fifty four (54) questions in Section B and four (4) questions in Section C.

3.4.2 Secondary data

Secondary date is the data that have been previously collected for some project other that then one at hand. (Zikmund, 2003). Secondary data is also known as historical data. Secondary data such as online journals, past research, and articles on organizational citizenship behaviours were accessed through Internet to assess further information for the research study.

All the secondary data were gathered and presented in Chapter two (2) literature review. The secondary data regarding organizational citizenship behaviour in restaurant industry were used to serve as evidence to support this research survey. In the process of doing this research, difficulty are encountered in distinguish between updated and obsolete information as obsolete data will cause the results of the research to be inaccurate. Hence, the data from articles and journals used were dated 2001 and onwards.

3.5 Questionnaire Design
3.5.1 Questionnaire Layout

According to Zikmund (2003), it is essentials to have a great layout and physical attractiveness questionnaire as it serves the purpose of gathering wide range of information from a large number of respondents. A relevance and accuracy questionnaire also will add merit for the researcher to obtain a good survey results. A presentable questionnaire will motive the respondent to cooperate from the beginning till the end of the questionnaire. There are various types of questions method such as scale questions, open ended questions and etc.

Scale Questions

Scale questions are used to collect attitude and belief data. (Saunders, Lewis & Thornhill, 2000). The responses from scale questions are graded on a continuum. For instance, rate the appearance of the product on the scale from 1 to 10, with 1 is the least preferred and 10 being the most preferred appearance. Types of scales include category scale, semantic differential scale, likert scale, and rank order scale. (Zikmund, 2003). In this research, likert scale will be used.

According to Zikmund (2003), likert scale measures attitudes designed to allow respondents to indicate how strongly they agree or disagree with a statement that ranges from very positive to very negative towards an attitudinal object. This type of scale can be found in four (4) questions in Section C of the questionnaire. For instance, the respondents are required to rate how important organizational citizenship behaviour to an organization on a scale that consists of strongly agree, agree, neutral, disagree and strongly disagree.

Open ended questions and closed ended questions

Open ended questions is a question that posses some problem and asks the respondent to answer in his or her own words. (Zikmund, 2003). Examples of open ended questions include: “What is your opinion on the questionnaires?” In this research, two (2) open ended questions can be found in Sections A and B of the questionnaire. For example, such as in question asking about race, there is a column provided for the respondents to fill in their answer if their answer is other than the three given options.

Whereas, closed ended questions are those which the respondent’s answers are limited to a fixed set of responses. There are several types of closed ended questions which include: Dichotomous questions whereby the respondents answer with a “yes”/”no” or “true”/”false” or “agree”/”disagree” response.

In this research, three (3) closed ended questions are being asked including gender, age, and educational level. For instance, in question asking about the gender of the respondents, only two (2) alternatives, male and female is being provided for the respondents to choose.

3.5.2 Measurement scales

There are four (4) types of measurement scales on the basis of the mathematical comparisons including nominal scale, ordinary scale, interval scale and ratio scale. All the scales have the same objective which is quantifying the information in order to coding, editing and analyzing data. The scales that used in this research are nominal, ordinal and likert scales.

Nominal Scale

Nominal scale is a scale in which the number assigned on a variable serves as labels for classification (Zikmund, 2003). In Section A of the questionnaire, demographic questions are asked in order to gather some basic information about the respondents. Most demographic variables are categorical variables and these variables are measured by using nominal scales. In this research, examples of demographic variables that use nominal scale are gender and course of study. For example, gender has two (2) categories which are male and female. Therefore number 1 can assign to male and number 2 to female. These two numbers are used to distinguish between the two categories.

For example:

1. What is your gender?

? Male ? Female

(b) Ordinal scale

Ordinal scale is a scale that arranges alternatives according to their magnitudes in an order relationship (Zikmund, 2003). In this research, examples of demographic variables that use ordinary scale are age and years of study.

For example:

What is your age?

? Below 18 ? 18 – 20 ? 21 – 23 ? 23 – 25 ? Above 25

(c) Likert Scale

In this research, likert scale is used in Section C. Section C of the questionnaire which consists of four (4) questions. In this section, likert scale is used to measure the factor which may influence employees toward organizational citizenship behaviours. The respondents were asked regarding the perceived organizational support, organizational commitment, organizational identification and organizational justice (distributive justice, procedural justice). The questions are measured in five (5) points including 1- strongly disagree, 2- disagree, 3- neutral, 4- agree and 5- strongly agree.

3.6 Pilot Test

According to Zikmund (2003), a pilot study is a small scale exploratory research project that uses sampling but does not apply rigorous standards. It is conducted to detect weaknesses in design and instrumentation (Copper and Shindler, 2003).

Twenty (20) sets of questionnaires were being distributed at random to respondents in any restaurants for pilot testing with 100% response rate which represent 13% of total sample size in this research. The research of pilot test is satisfying with the higher number of reliability test in SPSS. All calculated results are listed above 0.5 in reliability statistics. This shows that the pilot test was successfully carried out where the result of the reliability is under the “Good” category.

3.7 Method of Analysis

Data collected from the survey through questionnaire needs to be transformed into valuable information using the appropriate techniques. Statically package for Social Science (SPSS) techniques software of version 18.0 is used in this research. SPSS is considered user friendly which emphasizes on statistical calculations and hypothesis testing for varied types of data. The reliability test, frequency analysis, Pearson Correlation Coefficient Analysis, Independent Sample T-Test and Analysis of Variance (ANOVA) are being analyze using SPSS in this research.

3.7.1 Reliability Test

Zikmund (2003) define reliability as the measurements which are free from error and therefore yield consistent results. In this research, Coefficient of Cronbach Alpha was carried out to test for both consistency and the stability of the data collected to determine reliability. It reflects how well items in a set are positively correlated to one another (Sekaran, 2002).

3.7.2 Distributive Statistic
3.7.2.1 Frequency analysis

Frequency analysis is one of the descriptive analyses that make the researchers easy to understand and interpret the data collected. In frequency analysis, a set of data is organized by summarizing the numbers of times a particular value of a variable occurs (Zikmund, 2003).

3.7.3 Pearson Correlation

Correlation is used in this research to explore the relationships between two variables, and to determine whether the two variables have significant relationships, which are associated with the factors that affect successful of the company. The correlation is to measure the strength of linear relationship between two variables.

It is usually signified by (r), and can take on the values from -1.0 to 1.0. Whereas -1.0 is a perfect negative correlation, 0 is no correlation, and 1.0 is a perfect positive correlation. So, if the result falls into the range of -1.0 to 1.0, there is a significant relationship between both variables and should accept the alternative hypothesis (H1) and reject the null hypothesis (H0). However, if the result is equal to 0.0, there is no significant relationship between both independent variable and dependent variable should accept H0 and reject H1. (Hair, Money, Samouel & Page, 2007)

Table 3.1 Rules of thumb about Correlation Coefficient

Coefficient Range

Strength of Association

± .91 – ± 1.00

± .71 – ± .90

± .41 – ± .70

± .21 – ± .40

± .01 – ± .20

Very Strong

High

Moderate

Small but definite relationship

Slight, almost negligible

Sources: Hair, Money, Samouel & Page (2007).

3.7.4 Independent Sample T-Test

Independent sample t-test was used to analyze the data obtained from the questionnaire. According to Dr Shirley Holst (1999), independent sample t-test is concerned with testing from differences between data sets which come from different groups of individuals. For example, males versus females, or a group subjected to one manipulation versus another which acts as the control.

The objective in this research study is to test whether there is significant difference between the level of organization citizenship behaviours among different employees, the factor that influence employee to adapt to organizational citizenship behaviours and the level of association between employee characteristics and organizational citizenship behaviours. Thus, independent sample t-test will be a useful method to analyze the data.

3.7.5 Analysis of variance (ANOVA)

Analysis of variance (ANOVA) is a general method for studying the sampled data relationship. This method analyzes the difference between two or more samples, which is achieved by subdividing the total sum of squares (Shutler, 2002). ANOVA test is a technique to determine statistically significant differences between two or more groups. It is also a statistical technique for analyzing data to test the difference between two or more relationship by comparing the variances “within” groups and variances “between” groups (Merron, 2003).

3.8 Conclusion

This chapter discusses the methodology of research of the sampling design, data collection, questionnaire design and method of analysis. It also sets a questionnaire to test the relationship among the construct. Lastly it will proceed to the next chapter discussing about data analysis, finding and discussion to be used in this research.

Zikmund, W.G., 2003. Business Research Methodology. 7th ed., United States: South Western.

Saunders, M., Lewis, P., & Thornhill, A., 2000. Business Methods for Business Students. 2nd ed., Britain: Financial Times Prentice Hall

Clarke, G.M., and Cooke, D. 1998. A basic course in statistics. 4th ed. Arnold.