2. Step one: Major Points
1- How to enter data
2- What to compute? (PURPOSE)
3- Type statistics measurements (depends)
4- presenting data (graphs or tables)
5- Interpretation of data
3. Step two: Input & Output of new data
How to begin SPSS
:In your computer, SPSS uses two window pages
1-The data editor
2-The viewer
4. What is the data editor?
This is where you input your data and carry out all statistical
analysis. The data editor is composed of two pages:
1- Variable viewer (SPSS Viewer)
This is page where you identify your variables.
2- Data view
This is where you can see your data and compute it.
(Practice)
5. What is the Viewer?
This is where the results of your analysis appear
1- On the right hand side, there is a large space in
which the output is displayed.
2- SPSS displays both graphs and results on this page.
3- You ca edit graphs by clicking on the graph.
4- On the left hand side there is a tree diagram
illustrating.
(Practice)
6. Step three: How to enter data
1- You need to set the characteristics of each variable
by entering information into the columns
2- Information in the column include Name, Type,
Width, Decimals, Label, Values, Missing, Columns,
Align and Measure (details below)
7. Information in the column
1. 1-Names
1-Names must begin with a letter.
2-Names must not end with a period.
3-Names must be no longer than eight characters.
4-Names cannot contain blanks or special
characters.
5-Names must be unique.
6-Names are not case sensitive. It doesn’t matter if
you call your variable CLIENT, client, or CliENt. It’s all
client to SPSS.
8. Information in the column
2-Type: Use different types of data;
Dependent on variable type numeric (numbers)
/ string (containing letters)/ currency ($)
3-Width: preferred (8)
4-Decimals: select (2)
5- Label: restrictions in the characteristics of
variables
9. Information in the column
This column for assigning numbers to represent group
of people.
6- Missing
This column is for assigning number to missing data
10. Information in the column
7- Column
This for the width of the column. That is to say
how many characters should appear in the
column. This different from the width which
determines the width of the variable itself.
8- Alignment: left/ right/ centered
9- Measure: nominal/ ordinal/ scale
11. Hierarchy of measurement scales
• Nominal: category labels
– allows counting only (incidence, frequencies)
• Ordinal: rank order data
– distances between consecutive numbers are not necessarily equal
– arithmetic operations are meaningless
• Interval:
– real numners, but with arbitrary 0-point
– ratio's are meaningless
• Ratio:
– as interval, but with natural 0-point
– ratios are meaningful
12. String & Numeric variables
1- String variable
This consists of names, e.g. a student/ lecturer, job, income,
etc.
2- Numeric variable
This refers to variables you want view as numbers/ figures
Notice: if your data is words/ letters click on string variable
but if your data is numbers click on numeric variable
13. Step four: Descriptive statistics
Data analysis
Descriptive statistics:
– How to compactly summarize a lot of numbers
(measurements)?
See details below
14. Measures
• Central tendencies
– Mode
– Median
– Mean
• Excentricity
– Range
– Interquartile range
– Standard deviation & variance
15. Variables
• In terms of mathematical properties
– Discrete
• e.g. sex: male/female
• word length: 1, 2, 3, ... n letters
– Continuous
• There is always a value in between two other values,
e.g. vowel duration, age
16. Variables
• in terms of methodological function
– Independent: stimulus variable
– Dependent: response variable
The effect of X on Y
independent ~ dependent
19. Central measures
• Mode
– Most frequent value (or bracket of values) of a
variable
– Can be used even for nominal variables
20. Central measures
• Median
– The middle value after all scores have been
rankordered (or mid of two middle values if the
number of scores is even)
– Can be used for rankorder data or higher
– Cuts distribution into two equal halves (in terms
of surface)
– Relatively insensitive to skewed distributions
21. Step five: Inferential statistics
Inferential statistics:
– How accidental are the differences that I find
– What is the chance of drawing the wrong
conclusion?
22. Inferential statistics
• 2- Experimental studies
1-Independent variable under experimenter’s
control
2- Manipulation of reality in laboratory
3-Cause and effect can be distinguished
23. Inferential statistics
General strategy
1- Assume that the two samples have been drawn from the
same population
2- There should be no difference between the sample means
m1 = m2
3- This is called the null-hypothesis H0
4- If the chance of obtaining both samples from the same
population is less than 5%, then reject the null-hypothesis
and accept the alternative hypothesis H1