Introduction to Statistics - Summary, Tilburg University (7,5)

Frequency distribution: Analyze  Descriptive Statistics  Frequencies, enter variables
FREQUENCIES VARIABLES=…
/ORDER=ANALYSIS

Information about data: Analyze  Descriptive Statistics  Descriptives  Options adding mean,
standard deviation, variance, min, max.
DESCRIPTIVES VARIABLES =
/STATISTICS = MEAN STDDEV VARIANCE MIN MAX.

Histogram: Graphs  Legacy Dialogs  Histogram, Display Normal Curve, enter variable
GRAPH
/HISTOGRAM(NORMAL)=…

Scatterplot: Graphs  Legacy Dialogs  Scatter/dot  Simple Scatterplot  Define, put x-axis and
y-axis.
GRAPH
/SCATTERPLOT(BIVAR)=… WITH …
/MISSING=LISTWISE

Correlation: Analyze  Correlate  Bivariate, check pearson.
o Correlation significant means that we can conclude with CI certainly that there is a
relationship between the two variables. Also we are allowed to generalize to our finding to
the entire population our sample came from.
CORRELATIONS
/VARIABLES= … …
/PRINT=TWOTAIL NOSIG
/MISSING=PAIRWISE.

Outlier? Delete row or replace with missing value

Two groups: Data  Split File  Compare groups, enter variable
SORT CASES BY ….
SPLIT FILE LAYERED BY …

Sorting cases: Data  Sort Cases  enter variable
SORT CASES BY …

Only some cases: Data  Select Cases  If condition is satisfied  click “If..” and specify
COMPUTE filter_$=(…).
VARIABLE LABELS filter_$ 'hours … (FILTER)'.
VALUE LABELS filter_$ 0 'Not Selected' 1 'Selected'.
FORMATS filter_$ (f1.0).
FILTER BY filter_$.
EXECUTE

Computing variables: Transform  Compute Variable
COMPUTE …..
EXECUTE.

Recoding variables: Transform  Recode into Different Variables  output name, Change  Old
and New Values  specify recoding  Variable View  column “Values”  label values (also for
inverting scores)

, RECODE ….
(-10 THRU -1=1) (0-2) …
INTO ….
EXECUTE

VARIABLE LABELS …. “…”
VALUE LABELS … 0 “low” 1 “high”

Missing variable: Variable View  column “Missing” specify discrete missing value  recompute
frequency distribution
MISSING VALUES … (99)
EXECUTE.

Variable view
Name = short variable name
Label = longer description of variable
Values = meaning of values
Measure = level of measurement of variable

Percentage = dividing observed frequency by total N.
Valid percentage = dividing observed frequency by respondents with valid score.
Standard error of the mean = how much every sample differs from the mean of the population.

Describing normal distribution = the peak builds up but there is a drop in the middle, the extreme
scores have a low frequency. I would say this is not a normally distribution
Describing linear relation = it seems like there is a positive linear relationship between these two
variables; the lower the … the lower the …, and the higher the … the higher the ...

Clicking on “OK” will do it, clicking on “Paste” opens Syntax Editor
Always a period at the end of the command.

All options and command available: Help  Command Syntax Reference. Or selecting command in
Syntax Editor and then clicking on “Syntax Help”

Ending split file: command in Syntax Editor
SPLIT FILE
OFF.

Comments in Syntax Editor to know what you did:
COMMAND BLABLA
* This line is a comment.

One-sample t-test: Analyze  Compare Means  One-Sample T-test, enter variable
includes:
o N = sample size
o Mean = mean in sample
o Std. Deviation = standard deviation in sample
o Std. Error = standard error
o t = test statistic
o df = degrees of freedom
o Sig. 2-tailed = two-tailed p-value