Research methodology - Essay Example

Deciles divide the observations into ten equal parts and quartiles divide the observations into four equal parts. Measures of dispersion measures how varied the observations are in reference to several factors. This is an important factor in analyzing data in research because measures of central tendency are not enough or do not necessarily depict the data. Two data may have the same mean but have varying dispersions. That is why in reporting the mean, the minimum and maximum, range, standard deviation, and variance are reported as well.

The minimum and maximum values of course only reports the minimum and maximum observed values in the data thereby giving a picture of dispersion. The range reports the distance between the maximum and minimum values, which shows how wide or narrow the gap between these two values. For example, the observed heights of school children in one school varies from 47 inches to 55 inches (which are the minimum and maximum values) thereby giving a difference of 8 inches. In another school, the heights of the children vary from 46 inches to 57 inches thus the difference of 11inches.

Thus it can be said that the range of the heights of school children in the first school is less varied with that of the latter. Variance and standard deviation are both measures of how the observations vary in reference to the mean. If there is a large value for the standard deviation this means that the observations are highly dispersed and if the value is low, then the variation is less dispersed. In perception surveys, the variance and standard deviation are usually used to determine to what extent the subjects agree on their observations.

Conversion of data to and use of index numbers In as much as some quantities may not be directly observed given their values, they are converted to measurable value that can be manipulated for analysis. Index numbers on the other hand are numbers that measure relative changes of observations with reference to a base such that the raw data is not used in order to factor in changes in inflation and the like. The application of least squares regression analyses to data The main goal of a simple linear regression is to fit a straight line through the data that best predicts Y based on X.

If we want to know if age predicts IQ or if educational attainment predicts scores in information literacy, we use linear regression. The method of least squares is the one that minimizes the sum of the squares of deviations of the observed value of Y from its expected value. The calculation of correlation coefficients for data pairs; both Pearson’s product moment and Spearman’s rank correlation coefficients Correlations are used to determine the relationships of variables but not to predict.

For example, we want to determine the relationship between age and weight for employees of a particular firm, we analyze the data using correlation. In Pearson’s product moment correlation where the values range from 1 to –1 where the sign indicates the relationship. A positive sign shows a direct relationship while a negative sign shows an inverse relationship. If the correlation is +/- 1 this means that there is a strong relationship and low if otherwise. Spearman’s rank correlation involves ranking the values and is similarly interpreted like