CHAPTER 3 RESEARCH METHODOLOGY
This chapter presents the methods used in the study, the research instruments used in gathering the data needed, locale of the study, the respondents involved in the study, and finally, the statistical treatment of the data gathered.
Research Method Used
The researcher used the descriptive method of research. Descriptive Method is more expansive and encompassing than any other methods of investigation. In turn, descriptive method of research was das defined as: Descriptive research describes and interprets ‘what is’. It reveals conditions on relationships that exist or do not exist, practices that prevail or do not prevail, beliefs or point of view or attitudes that are held or are not held, processes that are going on or otherwise, affects that are being
Salvador, Baysa & Fua-Geronimo, 2008, felt or trends that are developing ( Salvador, p. 58). This method was utilized to help researcher portray the probable conditions of a particular situation. Also, descriptive method of research was used to describe characteristics of a particular population or phenomenon being studied; it does not answer questions about how/when/why the characteristics occurred rather, it addresses the “what” questio n. The reason for the widespread
use of this method is that it assists you in knowing how to accomplish your
29
desired purposes in the shortest available time. Descriptive studies have been used in many areas of investigation for the reason that they apply to varied kinds of problem. Thus, it is an appropriate method to be used by the researchers on their study in order to achieve the desire to improve the employees’ job satisfaction and performance, to effectively serve SSS members.
Respondents of the Study
The respondents of the study are employees of Social Security System (SSS) Cubao Branch which consists of 88 respondents.
Instruments Used
In this study, to facilitate tabulating and analyzing the data to be used and tested, a locally prepared closed form questionnaire was used to gather data on job satisfaction. The questionnaire was adopted from an unpublished undergraduate thesis entitled “Job Satisfaction and Performance of SSS Employees at Olongapo Branch Office ”; Santos Marilou Monteroso, (2003) on the same area but then was revised and modified to fit the current situation and environment of the subject under this study . This was used because it gathered data faster, easy to fill out, took little time, and kept the respondents on the subject matter and was objective than any other methods. Besides, the respondents are SSS professionals and so they were very literate. They can answer the questionnaire with ease.
30
It was composed of four parts: the first part dealt with Socio-Demographic Profile. This portion of the instrument focuses on the profile of the respondents, which includes name, age, gender, civil status, position, years in service, and monthly salary. The second part dealt with the organizational perspective of the SSS Cubao Branch, which includes type of organization, years of existence, number of employees, business hours, and days open. The third part dealt with the Level of Job Satisfaction. This part of the questionnaire was designed to gather information and examine the degree or level of satisfaction of the employees in the performance of their jobs. And the last part dealt with the cost of the benefits given to the employees to add to their satisfaction. A modified multiple-choice type of test was employed because of its simplicity, flexibility, and affectivity in measuring information. To assess the level of job satisfaction, the respondents were requested to check the space provided for each option for the degree of agreement on the stated questions using the Likert Scale which is describes as follows: WEIGHTED MEAN
SCALE
VERBAL INTERPRETATION
4.51 - 5.0
5
Strongly Agree
3.51 - 4.5
4
Agree
2.51 - 3.5
3
Undecided
1.51 - 2.5
2
Disagree
1.0 - 1.5
1
Strongly Disagree
31
Validity of the Instrument
Validity refers to the degree in which our test or other measuring devices is truly measuring what it intends to measure. For the validity of the instruments used, the researchers consulted their thesis adviser to check if the instrument can be used for the study. After the approval of their adviser, the researchers consulted the Branch Managers of SSS Cubao Branch to review the questionnaire if they will allow the researcher to distribute the questionnaires to the employees on their respective branches. The questionnaires was pre-tested and immediately edited before floating. After the consultation of the researchers from their adviser and the Branch Manager, the researchers performed the validity test. Validity test is the extent of which inferences, conclusions, and decisions made on the basis of test scores is appropriate and meaningful. It was conducted by the researchers to determine the degree to which an instrument measures what it intends to measure. For the validity of the instruments, the researchers used the item analysis. It is a procedure by which the validity index (VI) of each item in the test is determined. A decision was formulated as to which items or questions were to be retained, modified and discarded or replaced. Questions that resulted to a validity index of highly discriminating were retained, and those that resulted to moderately discriminating were modified. Conversely, a question that resulted to a validity index that was not discriminating was discarded.
32
The Validity Index formula is:
Where: = Percentage of the number of respondents who
answered Strongly Agree = Percentage of the number of respondents who
answered Strongly Disagree The succeeding classification is used to find out the level of validity of the test:
VI > 0.50
Item is highly discriminating
0.25 < VI < 0.50
Item is moderately discriminating
VI < 0.25
Item is not discriminating
In testing the validity of the survey questionnaire, the researchers made use of the ten samples given to the appropriate respondents of the study. The researchers first determine the proportion of the number of the respondents who answered “strongly agree ” and the proportion of the number of the respondents who answered “strongly disagree ” for each question then get their difference.
The difference was the validity index that was used to interpret the results. Most of the results obtained from the interpretation were highly discriminating, which indicates the validity of the questionnaire. To obtain the reliability of the entire test, a correction formula was applied, which is the Spearman-Brown Prophecy Formula, as follows: 33
To interpret the correlation coefficient value obtained, the following classifications may be applied: An r from 0.00 to ± 0.20 denotes negligible relationship An r from ± 0.21 to ± 0.40 denotes low or slight relationship An r from ± 0.41 to ± 0.70 denotes marked or moderate relationship An r from ± 0.71 to ± 0.90 denotes high relationship An r from ± 0.91 to ± 0.99 denotes very high relationship An r from ± 1.00 denotes perfect correlation The Pearson coefficient of reliability was computed as 0.82 and further measurement using Spearman Brown Formula resulted to 0.90 which denotes a high relationship.
Data Gathering Procedures
After securing the permission of the manager of the SSS Cubao Branch, the researchers distributed the questionnaires personally to the respondents. While conducting the survey, unstructured interviews were done by the researchers to obtain additional information that they could possibly use in the research. The researchers were very glad that they were able to retrieve one hundred percent of the questionnaires.
34
Statistical Treatment of Data
The raw scores from the questionnaire were tallied and tabulated for interpretation and analysis. Data cleaning was done after encoding data straight from the questionnaire. The researchers used the EXCEL and PASW Statistics software to facilitate the statistical computation of the data. The weighted values assigned to the statements were patterned after Likert Scaling: WEIGHTED MEAN
SCALE
4.51 – 5.00
5
Strongly Agree
3.51 – 4.50
4
Agree
2.51 – 3.50
3
Undecided
1.51 – 2.50
2
Disagree
1.00 – 1.50
1
Strongly Disagree
VERBAL INTERPRETATION
The Likert Scale method of Summated Ratings was used in the study. The Likert Scaling Technique assigns one scale value of each of the different responses. To analyze responses to a Likert Scale, each category is assigned to a numerical value such as Strongly Agree which is equal to 5 and Strongly Disagree which is equal to 1. The total assigned value will be determined by using the weighted mean. The scoring systems for each item must be such a high score consistently reflects a favorable response and a low score consistently reflects an unfavorable response. Likert Scales may be subject to distortion from several causes. Respondents may avoid using extreme response categories (central tendency); agree with statements as presented (acquiescence bias); or try to portray 35
themselves or their organization in a more favorable light (social desirability bias). Designing a scale with balanced keying (an equal number of positive and negative statements) can obviate the problem of acquiescence bias, since acquiescence on positively keyed items will balance acquiescence on negatively keyed items, but central tendency and social desirability are somewhat more problematic. The data gathered were analyzed statistically in order to answer some of the questions in this study and to be able to test the hypotheses set forth. The descriptive data were summarized taking their men, frequency, percentage and rank distribution. Job satisfaction was tested using one way ANALYSIS OF VARIANCE (ANOVA) to determine whether it changes significantly due to changes in the socio –demographic variables of the study. The weighted mean was computed in order to determine the overall perceptions of the respondents relative to their job satisfaction and performance. Percentage of that item is computed by dividing it with the sample total number of respondents who participated in the survey. The formula used in the application of this technique is:
P = ( ) x 100
Where:
P = Percentage f = Frequency
N = Number of classes or total sample After the researchers got the percentages of the item, the researchers used ranking. This is a descriptive measure to describe the numerical data in 36
addition to percentage. Ranking was used in the used for comparative purpose and for sharing the importance of items analyzed. It was used basically to determine the average responses of the different options provided in the various parts of the survey questionnaire used. The method was used in conjunction with the Likert Scale. To obtain the weighted mean, the following formula was used.
x= Where:
x = Weighted Mean The sum of all the products of f and x being the frequency of
each weight and x as the weight of each operation n = Total Number of Respondents
In determining the significant difference between two variables, the Analysis of Variance (ANOVA) was used. This tool is considered appropriate because ANOVA is one of the most widely and highly developed statistical methods in modern research. Wholey (2002) stressed further that technique of analysis is used whenever sample of data can be classified as groups. This technique provides the following information, first, it tests the different groups of data for homogeneity; second, it isolates and estimates the variations associated with defined sources. According to Davis (2002) it allows the sampling variation to be separated and their magnitude estimated. The following formulas are used to obtain the ANOVA:
37
a. SSb (between groups sum of squares) SSb
=
∑ [(n – Meant)2N]
b. SSw (within group sum of squares) SSw
=
[∑x1 – (∑x1)2 / N]
c. DFb (between groups degrees of freedom) DFb
=
K – 1
d. DFw (within group degrees of freedom) DFw
=
(N × K) – K
e. MSb (between groups mean sum of squares) MSb
=
SSb / DFb
f. MSw (within group mean sum of squares) MSw g. Fc where:
a.
n
=
SSw / DFw
=
MSb / MSw
=
average weighted mean
Meant =
average of the total average weighted mean
N
=
number of questions
K
=
number of choices
∑x1
=
sum of weighted mean
∑x12
=
sum of squared weighted mean
Hypothesis
Ho = There is no significant variation between the obtained/actual and expected frequencies. b.
Decision Rule
38
Reject Ho if computed X2 >X2.05; where X2 .5 critical value with Df.
39
