Data Analyse

ANALYSE DATA :  Finding and Conclusion

a)  DESCRIPTIVE

b)  INFERENCE

1. DESCRIPTIVE:  HOW TO RUN FREQUENCY COUNT for Nominal data

Step

1. Select Analyse  -> Descriptive Statistics   -> Frequencies

In Frequency dialogue box select e.g. Social demographic such as GENDER then click right arrow to move variable to the variable box

1. Select Statistics, tick Mean, Minimum, Maximum, then click Continue
2. Select Chart -> tick Bar Charts, then click Continue

Frequency count allow us to look at data structure and compute basic statistics about the distribution of the variable and it allows us to display data distribution in graphical form.

1. EXPLORATORY DATA ANALYSIS  for ordinal or interval or ratio data

Step

1. Select Analyse  -> Descriptive Statistics   -> Explore

In Explore dialogue box select Your DV and click the right arrow near the

Dependent List.

Under Display, select Both.

Select Factor List

e.g. Social demographic such as GENDER then click right arrow to move variable to the variable box

1. Select Statistics, tick Descriptive and Outliers in the Explore : Statistics dialogue box

Under Confidence interval for Mean, type 95, then click Continue.

1. Click  Plot  and under Box plots select Factor level together, then under Descriptive tick Stem and leaf and Histogram and Normality plots with test. then click Continue
2. Click Options and in the Explore: Option dialogue box, select Exclude cases listwise. then click Continue, select Ok.

Researchers compute descriptive statistics to describe the numerical characteristics of their data, however, descriptive statistics do not tell us  the underlying dynamics of the data.

To understand the  further into the relationships of variables, we need to compute inferential statistics. Inferential statistics use the law of probability to make inferences and draw statistics researchers to generalize  to a population  based on information obtained  from a limited number of research respondents.

Inferential statistics are important  because it is not possible in most instances to collect data  from populations. In order to make valid statistical estimations about populations, the researcher use RANDOM SAMPLES.

                                                     Random selection                      [pic 1][pic 2]

[pic 3]

[pic 4]

[pic 5]

[pic 6]

                                                               Inference

                                                             probability

The Chain of Reasoning for Inferential Statistics  (Wiersma, 2000) have 4 steps: