According to Merriam Webster online Dictionary (2009) methodology is the analysis of the principle of methods, rules and postulates employed by a discipline or a particular procedure or set of procedure. The systematic study of method that are, can be or have been applied within a discipline. The chapter research methodology describe the methodological approaches employed to test research hypothesis, the discussion is mainly focused on various aspects such a study design. Selection criteria for respondents, study sites, sampling procedures, and simple size, construction of measuring instruments, pilot study or pre- testing and measures adopted during development of questionnaire and statistical techniques used for data analysis such as uni- vitiate and bivariate analysis, chi square, gamma analysis are discussed.
“Any set of individual or object having some common observable characteristics constitute a population or universe” (Dixon and Marry, 1957). Universe is any set of individual or a subjects having common observable characteristics constitute a population or universe..
A sample is any subset of units form a population. A subset is any combination of sampling units that does not include the entire set of sampling sampling units. That has been defined as the population. A sample may be one sampling unit, all but one sampling unit, or any number in between (Nachmias and Nachmias, 1992). The factors of time cost and physical limitations usually play an important role in social researches. Therefore it is more economical and efficient to base studies on simples rather to study the entire universe
ossible technical modification in it, Experts’ viewpoint regarding the questionnaire ensure its content validity (Nachmias and Nachmias, 1192).
Quantitative data was analyzed through various statistics techniques. In univariate/ descriptive and bivariate techniques. In univariate analysis such as frequency, percentage and measures of central tendency (mean, standard devotion) were used to describe the data. In bivariate analysis, relationship among different variable was examined through applying chi- square and gamma tests. The relative important of independent variable in explaining dependent variable was based upon multivariate analysis.
USE OF SPSS
This revolutionary statistics analytical software system was called SPSS that stood for the Statistical Package for the Social Sciences. Nie, Hull and Bent (1968) developed SPSS out of the need to quickly analysis volumes of the social science data gathered through various methods of research techniques. The SPSS was used first time at Stanford University.
SPSS is the one of the most widely used software packages in the world of social sciences. It has been very crucial in facilitating the data of research using individuals as well as discrete units of analysis (Wellman, 1998). Father it is clear that SPSS technology has made difficult analytical targets easier by advances in usability and so data access and also enabling the researchers to benefit from the use of quantitative techniques in making decisions. It helps researcher to input the data on computer and can serve time from the laborious and exhaustive of an analysis.
It is defined as method for analyzing data on a single variable at a time. In univaraite analysis, each variable in any data set is explored separately. It also at the range of the values but as well as the central tendency measures of the values. It describe the pattern and tend of response to the variable, also describes the variable on its own. Descriptive/ univariate statistics interpret and summarize data also it describes individual variables.
Frequency distribution of different variables (Nominal variables Ordinal variables and interval variables) is done by identifying the lowest and highest values of the variable and then putting all the values of the variables in order form lowest to highest. Further, count the number of appearance of every values of the variable. It is a count of the frequency distribution with which each values occurs in the set of data. Frequencies distribution can also be presented in the form of percentage distributions and cumulative percentages. Descriptive data can be presented graphing, is a way of visual presenting of the data. Some researchers grasp the information into graphical form but it can be presented in a text format.
The mean is the most familiar average. It is defined as a obtained by dividing the sum of the all the observations. The mean gives indication of the magnitude of the observed values. The mean is usually denoted by placing a bar over the symbol used to represent the observation or the variable (Chaudhary, 1984). The mean was calculated with the following formula:-
X = the sum mean of a sample of size
? = Sum of observation
X =the responses obtained by all the respondents in a sample
N = Sample size
Bivariate analysis is applied to see the relationship/association between the two variables. It is explained in other words, which is explained by one variable is pattern in such a manner that is variables is not randomly distributed in connection with the other variables. Bivariate analysis was used to find out association of socio economic characteristics and the respondents and their decision making power. In its general form, relationship simply refers to the degree to which it because easier to predict value for the predict variables if one knows case’s value on the independent variable. This measure of association helps to understand relationship. Chi- square and gamma statistics were used to check the association and testing hypotheses of association and causality as well.
Chi –Square Test:
Chi-square these is depended by symbol x2Chi–square is defined as, a statistical test used to compare the observed with frequencies with expected frequencies, it would expect to obtain according to a specific hypothesis also to determine the degree of independence (Fisher, 1928). One of the most general and useful way to observe the information about the social globe in the set-up of the table, there are many ways to show information related to their question. The formula for calculating chi-square (X2) is:
X2 = ?(fo-fe)2 /fe
Chi-square is the sum of the squared difference between observed (fo) and the expected (fe) frequency (data) (or the deviation, d) and divided by the expected frequency in all possible categories. ?- capital sigma tells to compute the fractions for every cell and then sum over all cells to get x2. The following steps are involved to computer the chi-square:
i. To find out the difference between each observed frequency and the correspondence expected frequency for each cell in the table.
ii. Square for each difference.
iii. Divided each squared different by the respective frequency.
iv. Add the resulting division.
There are certain assumption of chi-square. At first, data is random sample of population, secondly data on nominal, ordinal. Thirdly there is not any expected frequency less than 5 and at lastly there is not any empty cell in frequency.
Gamma test is defined as, the strength of association/relationship of the crosstabulated data when two variables are being measured at the ordinal level of measurement (Sheskin, 2007) there values range from -1 means (100 percent negative association), or (perfect) to +1 means (100 percent positive association), or (perfect agreement), and value of zero shows the absence of association in the variables. This test is also known as Goodman and Kruskal’s gamma test). It’s very close to Someres’D and to Kendall’s tau.
In descriptive statistics, Gamma test statistic is an index of association between two variables which measure on ordinal levels, Suppose, if two pairs of scores are examined, they must either be concordant, in the sense that the one ranked higher than the order on the first variable is also ranked higher than the other on the second variable, or discordant,
The value of a gamma test statistic, (T) depends on two quantities:
Gamma = Ns – Nd
Ns + Nd
Formula shows that the size and the direction of gamma whether its positive, are functions of the relative number of same order as (Ns), versus order (Nd) pairs. However, more Ns pairs make the gamma positive and more Nd pairs make gamma negative and the larger the difference between Ns and Nd, the largest the size of the coefficient (irrespective of sign).
Ns, the number of pairs of cases ranked in the same order on both variables (member of concordant pairs),
Ns, the number of pairs of cases ranked differently on the variables (number of discordant pairs).
Gamma indicates the results in follow
i Gamma = Ns – 0 = 1.0
Ns + 0
A gamma of 1.0 indicates that the relationship between the variable is positive, and the dependent variable. When Ns is zero, gamma will be -1.0 indicating a perfect and a negative association between the variables
ii Gamma 0 – Nd
0 + Nd = -1.0
When Ns= Nd, gamma will equal zero:
iii Gamma = Ns – Nd = 0 = 0.0
Ns + Nd Ns + Nd
A gamma of zero reflect no association between the two variables; hence there is nothing to be gained by using the variables to predict order the dependent variable