You are on page 1of 19

Chapter 4

Data/Methodology
4.0 Tools of Data Collection: In order to find data on perceived organizational support and turnover intentions I will collect data on four variables: perceived organizational support (POS), organizational commitment (OT), job satisfaction (JS) and turnover intentions (TI) and for that I decided to use a questionnaire survey because it is a prevailing quantitative method used vastly by social science, management, and HRM researchers (Anderson, 2004). A survey describes a population (Sapsford, 1999) by providing a quantitative or numeric description of some fraction of the population the sample through the data collection process of asking questions of people this enables a researcher to generalize the findings from a sample of responses to a population (Creswell, 1994). In addition, questionnaire survey data can provide useful statistics to the researcher, which can be used to describe how things are and why the phenomena occur (Lewin, 2005). On the one hand, descriptive statistics can be used to describe and summarise data and include measures of central tendency (average) and dispersion (the spread of data or how close each other is to the measure of central tendency) (Lewin, 2005). On the other hand, inferential statistics are used to identify differences between groups, look for relationships between attributes and create models in order to be able to make predictions (Lewin, 2005). Therefore, by using a questionnaire survey method, the author was able to quantify answers to questions, compare and contrast responses among the samples look for relationships between variables, and enable generalizations from a sample of responses. More specifically, the present study will use a self-completion questionnaire, often referred to as a self-administered questionnaire referring to the type of questionnaire where respondents answer questions by completing the

questionnaire themselves (Bryman, 2001). The first benefit to use questionnaire technique is that It is good for measuring attitude and eliciting other content from research participants. It is Inexpensive. It can administer to probability samples. It can be quick turn around. It can be administered to groups. The Perceived anonymity by respondents possibly high. There is an ease of data analysis for close-ended items.

But on the other hand, there are some weaknesses of using questionnaire technique like

It needs validation. It must be kept short. The high ratio of non response to selective items The analysis of data sometimes time consuming for open-ended items.

Acknowledging that each method has some limitations but efforts will be made to minimize the weaknesses in order to increase the validity of the result. 4.1 Sampling Technique and Sample The Population of my study will be higher educational. Institutions of Pakistan in which the sample size of 100 teachers will be taken for conducting the study by using simple random sampling in order to generalize the findings in the particular sector. 4.2 Measures: In case of measuring Perceived Organizational support, Organizational Commitment, Job Satisfaction and turnover intentions only respondent filled questionnaire will be used. Participant

data was obtained thorough in one survey with 5-point likert scale only. Participant will asked a few questions. The survey contained a total of 28 questions and took approximately 10 to 15 minutes to complete. Incomplete survey was not included in the survey. There were different measures for the variables defined and were thought of with many complications. There were at least 5 questions in the survey obtaining data for one variable and some questions were directly obtaining data with no complications. Measure for each variable is defined below: Perceived Organizational Support Perceived organizational support was measure by asking the seven questions by using the 5points itemized rating type scale ranging from (1) strongly disagree to (5) strongly agree Organizational Commitment Organizational Commitment was measure by asking the seven questions by using the 5-points itemized rating type scale ranging from (1) strongly disagree to (5) strongly agree Job Satisfaction Job satisfaction was measure by asking the seven questions by using the 5-points itemized rating type scale ranging from (1) strongly disagree to (5) strongly agree Turnover Intentions Turnover Intentions was measure by asking the seven questions by using the 6-points itemized rating type scale ranging from (1) strongly disagree to (5) strongly agree 4.3 Methodology: To make analysis of data we will use SPSS software in which we will make analysis in to two parts where part one will lead descriptive statistics that will be use to describe and summarize data and include measures of central tendency (average) and dispersion (the spread of data or how close each other is to the measure of central tendency) (Lewin, 2005) and part two will

lead to inferential statistics that will be use to identify differences between groups, look for relationships between attributes and create models in order to be able to make predictions (Lewin, 2005). The methodology that we will adapt is discussed below one by one In order to get the overall summary of variable we will used descriptive statistics in which we will come to know about the nature of response that we will get from respondent like the value of maximum, minimum, mean and standard deviation. We will use frequency distribution which is a descriptive measure used for a graphical representation of the data to check the frequency (occurrence of one option) of the data as well as the normality curve to check either data is normally distributed or not. To check the overall picture for identifying the relationship between dependent and independent variable we will use scatter-plots. This matrix shows all the possible two dimension plots of the variables. Scatter plot will used to see the graphical picture of relationship between the dependent and independent variable. We will use regression analysis to find out the effect size of independent variable on dependent variable that how much change will occur in dependent variable due to change in independent variable by using regression equation for analysis. Where we will check the value of f-test to check the goodness of fit of the model. We will also make focus on the value to T to check the significance level of relationship for the acceptance or rejection of our alternative hypothesis. We will also make focus on the value of R square to see the contribution level of variable mean how much independent variable contribute in changing the value of dependent variable. Since the objective of this study is to check the impact of perceive organizational support on turnover intentions in higher educational sector of Pakistan. So, here in this study ordinary least square (OLS) method of simple regression will be used to find such equation which could be used to

find the impact of perceived organizational support on turnover intentions and the specified regression equation will take the following form POSi = C + B(TIi) + Ei The equation which shows above where POSi is the dependent variable and C is the value of constant, B(TIi) is the value of independent variable shows the partial regression coefficient represents the change in dependent variable, due to one unit change in independent variable. Ei is the error term. We will also use correlation to find the relationship among variables as well as check the strength of the relationship of different variables. It will also show the direction of relationship between two variables.

Chapter 5
Analysis
The result of the study analysis is presented in this chapter. Descriptive analysis used to describe the data by using descriptive summary as well as histogram with normal distribution curve. Inferential analysis used to describe the relation between variables by checking the acceptance or rejection of hypothesis and to see the nature of relationship between variables. In inferential

portion the study tested the relationship between perceived organizational support (POS), job satisfaction (JS), organizational commitment (OC) and turnover intentions (TI). 5.1 Data Screening: Data screening and preparation is focused on ensuring that the data contained the appropriate range of scores for all analyzed variables including the identification of possible outlines. We used questionnaire for collection of data. After that the most complicated work is to make analysis of the data. So, for that purpose the first step is data screening in which we ensure that all the participants were filled the questionnaire correctly and completely and after that we coded all returned questionnaires in to Spss software for data analysis. For every variable we used 7 questions so, firstly we compute all the variables to get the final shape of our data variables and after that used different test for analysis. Lets discussed the detail of each variable one by one in the results portion. 5.2 Results: In this portion all the results with their interpretation were shown and being discussed one by one

5.2.1 Descriptive Summary

Table 1.1
Descriptive Statistics N pos commitment satisfaction Turnover intentions Valid N (listwise) 100 100 100 100 100 Minimum 2 2 2 2 Maximum 5 5 4 5 Mean 3.63 3.43 3.31 2.75 Std. Deviation .537 .479 .492 .622

Table 1.1 presents the descriptive statistics that show the overall picture of all the four variables. There were scales of 5 responses that lead to the options (strongly disagree, disagree, neutral, agree, and strongly agree). Number of observations of each variable is 100. In the above table the mean values and the values of standard deviation of all the 4 variables have been shown. Mean value provides the idea about the central tendency of the values of a variable. For example if we observe the above output to assess the average response rate or the respondent then we come to know the mean of different variables like POS (mean:3.63), commitment(mean:3.43), satisfaction (mean:3.31) and turnover intentions (mean: 2.75). If we observe then for the first three variables (POS, commitment and satisfaction) the average response rate of responded is lie with in the option 3-4 (3 is for neutral and 4 is for agree) and for turnover intention the average response rate of responded is lie in the option 2-3 (2 is for disagree and 3 is for neutral). The minimum option that is ticked by responded is 2 and the maximum option that is ticked by responded is 5. Standard deviation gives the idea about the dispersion of the values of a variable from its mean value. So, if we observe then in the response rate for the variable of commitment is value of standard deviation is (S.D.479) which is the lowest value as compare to other variable values. Which shows that most of the respondent answers were same for the variable of commitment and

have consistency in their response rate but if we observe then for turnover intention the value of standard deviation is (S.D .622) which is quite high as compare to other variables which clearly shows that the response regarding turnover intention of mostly respondents were not the same and they dont have consistency in their answers.

5.2.2 Histogram This shows the graphical representation of the variables with the curve to check the normality of the response rate. Lets discuss the result of each variables histogram one by one Figure 1.1

Strongly Disagree Disagree Neutral Agree Strongly Agree

= = = = =

1 2 3 4 5

The fig.1 shows the graphical representation of the bars that is showing the response of the respondents regarding perceived organizational support. Most of the participants lies in the option 3 - 4 (3 is for neutral and 4 is for agree). Similarly small numbers of respondents were marked very low and very high options. The bars in the histogram from a distribution (pattern or curve) that is similar to the normal, bell shaped curve. Thus, frequency distribution of the perceived organizational support is normal. Figure 1.2

Strongly Disagree Disagree Neutral Agree Strongly Agree

= = = = =

1 2 3 4 5

The fig.1 shows the graphical representation of the bars that is showing the response of the respondents regarding organizational commitment. Most of the participants lies in the option 3 4 (3 is for neutral and 4 is for agree). Similarly small numbers of respondents were marked very

low and very high options. The bars in the histogram from a distribution (pattern or curve) that is similar to the normal, bell shaped curve. Thus, frequency distribution of the commitment is approximately normal.

Figure 1.3

Strongly Disagree Disagree Neutral Agree Strongly Agree

= = = = =

1 2 3 4 5

The fig.1 shows the graphical representation of the bars that is showing the response of the respondents regarding job satisfaction. Most of the participants lies in the option 3 - 4 (3 is for neutral and 4 is for agree). Similarly small numbers of respondents were marked very low and very high options. The bars in the histogram from a distribution (pattern or curve) that is similar

to the normal, bell shaped curve. Thus, frequency distribution of the job satisfaction is slightly skewed but approximately normally distributed. Figure 1.4

Strongly Disagree Disagree Neutral Agree Strongly Agree

= = = = =

1 2 3 4 5

The fig.1 shows the graphical representation of the bars that is showing the response of the respondents regarding turnover intentions. Most of the participants lies in the option 2 - 3 (2 is for disagree and 3 is for neutral). Similarly small numbers of respondents were marked very low and very high options. The bars in the histogram from a distribution (pattern or curve) that is similar to the normal, bell shaped curve. Thus, frequency distribution of the intentions is approximately normal. 5.2.3 Scatter Plots

Scatter plot or graph of two variables shows how the scores for an individual on one variable associates with his or her scores on the other variable. Lets discuss each plot or graph one by one Figure 1.1

Figure 1.1 shows the results of scatter plot matrix where we intend to have some idea about the relationship between perceived organizational support and commitment. If we observe then the flow of line is come from right to left which shows the positive relationship between perceived organizational support and commitment. This means that if the organization gives high perceived organizational support then there is also increase in the commitment level of an employee and the above results have been confirmed by the table of correlations. Figure 1.2

Figure 1.2 shows the results of scatter plot matrix where we intend to have some idea about the relationship between perceived organizational support and satisfaction. If we observe then the flow of line is come from right to left which shows the positive relationship between perceived organizational support and satisfaction. This means that if the organization gives high-perceived organizational support then there is also increase in the satisfaction level of an employee and the table of correlations has confirmed the above results. Figure 1.3

Figure 1.3 shows the results of scatter plot matrix where we intend to have some idea about the relationship between commitment and satisfaction. If we observe then the flow of line is come from right to left which shows the positive relationship between commitment and satisfaction. This means that if the commitment level of an employee is increase then there is also increase in

the satisfaction level of an employee and both variables are mutually correlated with one an other and the table of correlations has confirmed the above results.

Figure 1.4

Figure 1.4 shows the results of scatter plot matrix where we intend to have some idea about the relationship between commitment and turnover intentions. If we observe then the flow of line is come from left to right that shows the negative relationship between commitment and turnover intentions. This means that if the commitment level of an employee is high then there is low turnover intentions occur for an employee and the above results have been confirmed by the table of correlations.

Figure 1.5

Figure 1.5 shows the results of scatter plot matrix where we intend to have some idea about the relationship between satisfaction and turnover intentions. If we observe then the flow of line is come from left to right that shows the negative relationship between satisfaction and turnover intentions. This means that if the satisfaction level of an employee is high then there is low turnover intentions occur for an employee and the above results have been confirmed by the table of correlations. Figure 1.6

Figure 1.6 shows the results of scatter plot matrix where we intend to have some idea about the relationship between perceived organizational support and turnover intentions. If we observe then the flow of line is come from left to right that shows the negative relationship between

perceived organizational support and turnover intentions. This means that if the perceived organizational support level of an employee is high then there is low turnover intentions occur for an employee and the above results have been confirmed by the table of correlations. 5.2.4 Correlation Correlation is used to check the mutual relationship among variables. For checking the relationship we will make two hypotheses: null (H0) and alternative (H1). We interpret the findings on the acceptance or rejection of the hypothesis. We used correlation matrix to check the mutual relationship of different variables. The hypothesis which we developed are given below Hypothesis 1: H1: there is relationship between POS and organizational commitment H0: there is no relationship between POS and organizational commitment Hypothesis 2: H1: there is relationship between POS and H0: there is no relationship between POS and job satisfaction Hypothesis 3: H1: there is relationship between organizational commitment and job satisfaction H0: there is no relationship between organizational commitment and job satisfaction Hypothesis 4: H1: there is relationship between organizational commitment and turnover intentions H0: there is no relationship between organizational commitment and turnover intentions Hypothesis 5: H1: there is relationship between job satisfaction and turnover intentions H0: there is no relationship between job satisfaction and turnover intentions Hypothesis 6: H1: there is relationship between POS and turnover intentions

H0: there is no relationship between POS and turnover intentions


Table 1.1 Correlations pos Pos Pearson Correlation Sig. (2-tailed) N commitment Pearson Correlation Sig. (2-tailed) N satisfaction Pearson Correlation Sig. (2-tailed) N Turnover intentions Pearson Correlation Sig. (2-tailed) N 100 .422
*

commitment Turnover intentions satisfaction 1 .422


*

-.201

.592 .000 100 .792


**

.000 100 1

.045 100 -.232 .042

.000 100 .592 .000 100 -.201


*

.000 100 1

100 .792
**

100 -.212

.000 .032 100 -.232 .042 100 100 100 1 .032 100 100 -.212

.045 100

Table 1.1 represents the table of correlations.

Where two variables commitment and

satisfaction are positively correlated to perceived organizational support (r= .422, p = .000, and r= .592, p= .000, respectively). Turnover intention is negatively correlated to the perceived organizational support (r= -.201, p = .045). The magnitudes of the above discussed two correlations are greater than 0.33 in the absolute terms, which shows the moderate correlations between the said pairs of the variables but the correlation of turnover intention is lesser than 0.33 in absolute terms, which shows the weak correlation between POS and turnover intention. If we see the mutual relationship of commitment and satisfaction (r= .792, p = .000) then there is positive relationship exist between both variables and the value of r= .792 which shows that both variables are strongly related to each other and the variables commitment and satisfaction are

negatively correlated to turnover intentions (r= -.232, p = .042, and r= -.212, p= .032, respectively) but both correlation were lesser than 0.33 which shows that both have weak impact on turnover intentions. All the above correlations are statistically significant at less than five percent level of significant. In the case of these correlations the null hypothesis that were stated above of no correlation are rejected as the P-values are lesser than 0.05. 5.2.5 Regression Regression is used to check the effect size of independent variable on dependent variable. Results of regression analysis are given in table 5.3. The value of the coefficient of determination (R2) is .167. This shows that the correlation between the observed values of perceived organizational support and the fitted value of the turnover intention is 16% percent. The adjusted coefficient of determination (adj. R2) shows is adjusted for the degrees of freedom. The value of the adjusted coefficient of determination (adj. R2) is not affected. The value of the adjusted coefficient of determination (adj. R2) is .158, which shows that 15% variations in turnover intentions. The value of F-statistic is statistically significant at less than five percent that exhibits that in the estimated model at least one of the partial regressions coefficients is different from zero.
Table 5.3 Coefficients
a

Standardized Unstandardized Coefficients Model 1 (Constant) pos B 1.031 .473 Std. Error .391 .107 .409 Coefficients Beta t 2.633 4.432 Sig. .010 .000

a. Dependent Variable: turnover intentions

Necessary Statistics R2 = 0.16 F = 19.748 SigF = 0.001

T = 1.03 + 0.47 (POS) (2.63)* (4.43)*

----------- EQ 5.11

The coefficient table presents the results of the regression analysis. The objective of the regression in this study is to find such an equation that could be used to find the impact of perceived organizational support (POS) on turnover intentions (TI). The specified regression equation takes the following form: The results show that the independent variables less significantly affect the employee turnover intentions as shown by the values of the t-statistic and the corresponding P-values. T-test is used to test the significance of the individual partial regression coefficients. Null hypothesis in this test is set as the partial regression coefficient is zero. This test shows that the coefficients of the predictor are statistically significant at less than five percent level of significance.

Value in parenthesis are t ratios and ***, ** and * represent significance level at 10 %, 5% and 1% respectively.

You might also like