Location basically refers to the statistical estimation of the provided data in order to get a central value (Salvatore & Reagle 2002, p. 25). The central value also known as the typical value is defines in various ways.
The mean refers to the sum of the data points provided divided by actual number (N) of the data points. The formula is given as follows:
The mean is common used as statistical estimation because its takes the true value into account. However; in case where they are extreme value in data points provide it gives an inaccurate estimation.
The median is very simple, easy to calculate and can sometimes obtained by simply inspection. It is the value of the point which has half the data. The formula definition is as follows:
Unlike the mean the median lies in the middle part and hence is not affected by extreme data points. However; the median sometimes gives an accurate representation of the actual average since it does not depend on the data points in the series.
Mode is the value which appears most frequently in the data provided. Therefore; helps on to see what happnes on theall data set. Like the median its easy, quick and simple. However; is very unrelaible to find the average and the middle of the data set.
The difference between the highest value and the lowest is called the range.
Advantages: Fast data analysis; the exact scores are retained; and helps show both mi9nimum and maximum. Disadvantages: not suitable for wide range of data; it is suitable for scores that are less than 50; and it isn’t appealing visually.
It is the difference between the 75th percentage mark and the 25th mark. Advantages: when comparing data, it is simpler to use than the range; and doesn’t make any use of extreme scores.
The standard deviation is the square root of the variance.
Advantages: Since it is not expressed in squared units, it gives sensible answers; much data within a single standard deviation that is either over or lower than the mean. Disadvantage: it is affected by extreme scores.
Index numbers are numbers that are planned to assess the scope of of changes that happen economically in a period of time. Advantages: Assists in comparisons; it facilitates assessments of either percentile changes or relative changes and not absolute changes; and easier access to the statistics since there are accessed sequentially. Disadvantages: since there are varying types of index numbers the one to be applied depends on the situation; and there is loss of data since a single index signifies a lot of data.
Correlation and regression
This is a method of describing the relation between predictor variables and other variables.
Advantage: it may predict the scores for the dependent variable in situations where the only known factor is the independent variable. Disadvantage: It has restrained academic value; they don’t zap a causality clue; the calculation doesn’t let the researchers known the association between the data set that is accountable for the correlation equations statement.
This is a test that is carried out to assess whether there is a dramatic dissimilarity between the anticipated frequencies and those that are observed frequencies.
Advantages: Can be used to test the relationship between variables; and lets the user the dissimilarity between both the anticipated and observed results. Disadvantages: the percentages are not seen; has limitations in that data must be in numerals and groups; and the number to be observed mustn’t be less than twenty.
They are used to test the statistical difference of means between two or more groupings.
Since it requires a little space and equipment, it becomes a simple test to carry out. Disadvantage: the test can only be performed by none person at a given time.
In a research, the measures are used:
In a research setting, the measures are used: means (assess the statistical quantity), standard deviation (know the varieties to occur), chi-square (helps to determine the significance), and know the solve other basic existing problems. Also with tradition research methods, it would be difficult to conduct researches. Using these researches, gives timeliness compared to traditional methods.
In research techniques: the internal methods that are consistent; they assist to get data that is relevant according to the objectives of the research; and the assist in getting data that is reliable and valid when it is compared to other traditional ways.
Dispersion has been said to be the spread of values around a central tendency. In dispersion, measurements are in the form of the range and the standard deviation.in order to get the range, the diffrence between the highest values and lowest values is sort (Salvatore & Reagle 2002, p. 40). The standard deviation finds the relation of the points provided in data to the mean.
Index numbers measure the changes that occur over time to the data. It is similar to getting a percentage. The formula involves having the most recent digits over the previous digits then multiply by a hundred. When dealing with the matter of least square regression analysis it is important to know what relationship exists between the dependent variable and independent variables? This is normally used when making a prediction.
A correlation occurs when one of the figures in the data provided varies directly with another. Correlation coefficient determines the extent to which the variables relate to each other (Salvatore & Reagle 2002, p. 45). Pearson discusses calculation of the correlation coefficient using the following formula:
r is the mean of the sum of the products and is the variable of the two variables.
Spearman’s different formula for calculation is presented as follows:
D= Rank of X – Rank of Y (i.e., a difference score).
Using the chi square method of analyzing data will help in testing for a null hypothesis. The data in this kind of a situation is made up of discrete variables. The t test is used when making a comparison of means from two different sources of data. This is in the relation to the variation in data. The t value is calculated using the following formula:
Salvatore, D & Reagle D.P 2002, Schaum’s outline of theory and problems of statistics and econometrics, 2nd Edition, McGraw-Hill Professional, New York