Points to remember

 

Points to remember

 

<< back to chapter contents

The following are the major points introduced in this chapter. Ensure that you are very confident with their meaning, content, context and significance.

 1     Tables and graphs are the most popular methods of data presentation in qualitative research.

 2     Table presentation is expected to adhere to rules of clarity, simplicity, economy of space, order of variables, appearance, accuracy and objectivity.

 3     Graphs present data visually. Most graphs are constructed within the framework of the coordinate axes; the X-axis (abscissa) and the Y-axis (ordinate).

 4     A mean describes the central trend or average of all observations.

 5     Mode is the category of a distribution that has the largest number of observations.

 6     Median is the point on a distribution that divides the observations (not their values) into two equal parts.

 7     The variance is the average of the squared deviations from the mean.

 8     Standard deviation is the square root of the variance.

 9     The range describes the distance between the lowest and the highest score in a distribution.

10    Correlations display the relationship between two variables.

11    Correlation coefficients demonstrate the presence or absence of correlation, the direction of correlation and the strength of correlation.

12    A correlation coefficient ranges from +1 to -1.

13    Spearman's rho and Pearson's r are the most commonly used measures of association.

14    Spearman's rank order correlation coefficient is employed when data are ordinally measured.

15    Pearson's rank order correlation coefficient is suitable for interval/ratio measured data.

16    A positive correlation means that when one variable is increased the other will increase too.

17    A negative correlation indicates that an increase in one variable is associated with a decrease in the other.

18    A zero correlation means that there is no association between the variables.

19    The coefficient of determination is the square of the coefficient of correlation and displays the degree of variability shared by the two variables.

20    Tests of significance relate statistics to parameters; the sample to the target population; the study to the society.

21    Parametric tests assume that the variable in question is normally distributed in the population; they employ the normal curve.

22    Non-parametric tests do not assume that the variable is normally distributed in the population.

23    Tests of significance work around the notion of null hypothesis (Ho): if the significance level of a particular test is below .05, the Ho is rejected; if it is above .05, the Ho is accepted.

24    The most common tests of significance are chi-square test, t-test and analysis of variance.

25    Chi-square tests are suitable for nominal data; t-tests and ANOVA are suitable for interval/ratio level data.

26    The t-test and ANOVA compare means rather than scores. They answer the question whether the differences between the means are significant or not.

27    Computer programs such as SPSS can assist with computing significance tests, making relevant computations and conclusions fast, precisely and easily.




Workbook Home

Preface | Introduction | Varieties of social research | Feminist research | Principles of social research | Research design | Initiating social research | Sampling procedures | Multi-sample studies | Field research | Observation | Surveys: questionnaires | Surveys: interviews | The study of documents | Applied research | Qualitative analysis | Quantitative analysis | Reporting

Copyright © Sotirios Sarantakos