Chapter 4 - Statistical analysis Flashcards
What are the major steps to prepare data for statistical analysis?
* Clean the dataset.
Coding ( put numbers on the data )
Organize to software
What are the measures of central tendency?
Median: midt i rekkefølge
Mode/modus: Det som forekommer mest
Mean: gjennomsnitt
What are the measures of spread?
Range: Max - min.
Variance: Avstanden fra gjennomsnitt opphøyd i 2, sammenlagt og tatt gjennomsnittet av.
Standard avvik: kvadratrot av varians.
(Høye verdier = stor spredning)
What is normal distribution? Why is it important to test whether a data sample is normally distributed?
Normal distribution is to distribute the scores normally. It must meet a requirement for parametric tests to be appropriate. (s.76)
What statistical methods are available for comparing group means?
T-test and ANOVA.
T test gets a T value. If bigger the bigger chance there is for the nullhypothesis to be
false.
Also generates a P value that says something about the chance of the answer being a
coincidence. (should be under 5%)
What statistical method can be used to compare two group means contributed by two
independent groups?
Indipendent samples T-test: betweengroup. (s.75-6)
What statistical method can be used to compare two group means contributed by the same group?
Paired sample T-test: withingroup.
s.75-6
When should a oneway-ANOVA test be used? Describe a research study design that fits the oneway ANOVA test.
Basic design with 1 variable and more than 2 conditions.
Betweengroup.
When should a f actorial ANOVA test be used? Describe a research study design that fits the factorial ANOVA test.
Factorial design (>1 variable). betweengroup.
When should a repeated measures ANOVA test be used? Describe a research study design that fists the repeated measures ANOVA test.
Within-group, more than 1 IV or more than 2 conditions.
When should a s plitplot ANOVA test be used? Describe a research study design that fits the splitplot ANOVA test.
Factorial design.
Splitplot.
When should correlation analysis be used? What does Pearson’s r square represent?
Er det et statistisk forhold mellom 2 variabler?
X påvirker Y. Når man har gjort et korrelasjonsstudie. Pearsons R sier noe om styrken til forholdet. (1 til 1)
0 = ikke noe forhold.
-1 = negativt forhold.
1 Positivt forhold.
R square sier % av X som blir påvirket av Y. (0.5^2 = 25%)(s. 86).
When should regression analysis be used? Describe a research study that requires regression analysis.
Ligger under korrelasjonsanalyse.
Forholdet mellom 1 og flere andre.
Prediksjonssvar = en uavhengig variabel blir brukt for å forutse et annet svar.
Name two regression procedures and discuss when a specific procedure should be used.
Sjekk boka
What are the assumptions for parametric statistical tests?
Tester som har en normalfordeling.
Ikkeparametiske tester har ikke en normalfordeling.
Sier om det er en normalfordeling eller ikke.
Sier ikke mye mer.
