characteristics and will serve as a guide to the explanations which follow 78

Characteristics and will serve as a guide to the

• 117

This preview shows page 77 - 80 out of 117 pages.

characteristics, and will serve as a guide to the explanations which follow:
78 The major classes of statistics are parametric and non-parametric statistics. An understanding of the meaning of a parameter, which in this context refers to a function of the population, is essential in order to appreciate the difference between these two words. A parameter of a population is a constant feature which it shares with other populations: a common one is the ‘bell’ curve of the normal frequency distribution. Most populations display a large number of more or less ‘average’ cases with extreme cases tailing off at each end. For example, most people are of about average height, with those who are extremely tall or short being in a distinct minority. The distribution of people’s heights shown on a graph would take the form of the normal or “Gaussian curve shown below. Although values vary from case to case, the generality of this type of curve amongst populations is so strong Inferential StatisticInference Predictions Hypothesis testing Estimations Variation Correlation Central Tendency Parametric Statistics Non-Parametric Statics Randomness Variance Association Correlation Distribution STATISTICS
79 that statisticians take it as a constant – a basic parameter – on which the calculations of parametric statistics are based. For those cases, where this parameter is absent, non-parametric statistics may be applicable. 3.4Descriptive Statistics The two classes of parametric statistics are descriptive and inferential statistics. Descriptive statistics provide a method of quantifying the characteristics of the data, where their centre is, how broadly they speed and how one aspect of the data relates to another aspect of the same data. The “centre of gravity’ of the data, their point of central tendency can be determined by finding the ‘mode’ or the ‘median’ and any one of several ‘means’. These measures have their own characteristics and applications and should be chosen with regard to the data being analyzed. Frequency count Mean Score Mean and Mode coincide with the mean The measure of the dispersion (or spread) of the data, how flat or steep the Gaussian curve appears, is an indication of how many of the data closely resemble the mean. The flatter the curve, the greater is the amount of data that deviate from the mean i.e. the fewer that are close to the average. The horizontal length of the curve also gives an indication of the spread of values and the extent of the extremes represented in the data, while the occurrence of a non-symmetrical curve indicate skewness in the data values. Apart from examining the qualities of a single set of data, the main purpose of statistical analysis is to identify and quantify relationships between variables. This is the type of research called correlation research. But remember, the mere discovery and measurement of correlation is not sufficient on its own to provide research answers. It is the interpretation of