Measures of central tendency
Central tendency is a term in statistics indicating how the data gathers around a given value. The data in question is quantitative data. The specific value where the data clusters is what is known as the central value. There may be more than one value around which quantitative data may cluster. For purpose of convenience, these values are given specific identity. The different means in which this specific value is identified is referred to as the measures of central tendency. They focus on the typical value of a data set. According to Howell (2012), the row data need to be displayed in a manner that the analyst will be able to draw conclusions about it. For instance, the data may be plotted so that its distribution shape may be viewed. The statistics that are used to represent the center of distribution are referred to as measures of central tendency. They are sometimes referred to as measures of location and represent a set of measures that indicated where on a scale the distribution is centered.
The measures of central tendency differ in how much use they make of the data, especially of the extreme values, but they are all trying to explain something about where the center of distribution lies. The three major measures of central tendency are the mean, mode, and median, and each of them explains something unique about the data. According to Leach (2004), the mean represents the numerical average of a set of values. For instance, consider the following set of data: (2, 3, 5, 4, 3, 2, and 2). The mean for these values is (2+3+5+4+3+2+2)/7 = 3. This shows that 3 is the mean of this data set. The median is the figure that lies at the middles of a data when it is arranged in ascending or descending order. In our case, the median for (2, 2, 2, 3, 3, 4, and 5) is 3. The mode is the figure that is appearing most frequently in a given data set. The mode of the data above is 2 because it is appearing three times is the data set.
Measures of variability
Measures of central tendency are not the only measures used to make inferences about a frequency distribution. There is a need to consider the spread of a distribution. This can only be achieved by use of the measures of central tendency. According to Black (2011), measures of variability indicate how the elements of a data set are spread. The measures of central tendency do not provide information about the spread of data values. This information is available in the case of measures of variability. The measures of variability indicated how widely the observations are spread out around the measures of location. Variance, range, and standard deviation are the commonly measures used to assess how observations are spread. They increase with increase in variation of variables. In case there is no variation, the measures of spread will be equal to zero.
Range is the simplest measure of spread. The range measures or represents the difference between the largest and the smallest figure in a data set. The range is the easiest to compute but it has limited usefulness. The difference between the first and the third quartile is referred to as the interquartile range. It is computed by getting the range of the middle 50% of a data set. The range is affected by the extreme values. When there are too small or very large extreme values, the range will be affected. The range is not sensitive to other values because it only uses the lowest and the highest value. It is not reliable as a measure of spread because it does not take into account all the other figures in the data set.
The deviations from the central tendency squared and averaged should give a researcher the variance of his data. When the observations in the distribution are widely spread, the variance will be large and vice versa. The standard deviation is computed by finding the square root of a given variance.
Measures of relationship
Measures of relationship try to explain any relationships that occur between two or more variables. Correlation coefficients are uses to assess the extent to which variables in a data set are related. The manner in which the variables tend to vary together is called relationship between variables. The relationships between variables are obtained using such methods as correlation analysis, scatter plots and linear relationship. The graphs may be used to explain the trend that exists between two or more variables (Osborn, 2006). The two major methods that are used to test for the relationship between variables include the spearman Rank – order correlation, and Pearson Product-Moment Correlation. The simple regression analysis is also helpful in analyzing how one variable, called the independent variable cause the other, the dependent variable. Multiple regression analysis is also used to explain how several variables jointly cause the other.
Reference List
Black, K. (2011). Business Statistics: For Contemporary Decision Making. UK: John Wiley & Sons.
Howell, D. (2012). Statistical Methods for Psychology. New York: Cengage Learning.
Leach, R. (2004). The Chiropractic Theories: A Textbook of Scientific Research. London: Lippincott Williams & Wilkins.
Osborn, E. (2006). Statistical Applications for Health Info Mgmt 2e. London: Jones & Bartlett Learning.