Statistics Flashcards
(8 cards)
Descriptive versus Inferential Statistic
Descriptive statistics(1) Central tendencyMeanMedianMode
(2) Measures of SpreadVarianceStandard Deviation
Descriptive statistics can describe a data setBUT … they do not allow you to say anything about people who were not included in your data setI want to know the mean GPA of U3 psych majorsI don’t have time to ask everyoneSo I ask 20 people*Mean of 20-person sample = 3.3
Sample versus populationI want to know if U3 psych majors at McGill have higher GPAS than U3 economics majorsIf I obtain the GPA of every single U3 econ and psych major, I cancompare the meansIf I obtain the GPA of a subsetof U3 econ and psych majors, I cannotjust compare the means
Inferential StatisticsSample 20 U3 psych majors and 20 U3 econ majorsPsych majors – mean GPA is 3.0Econ majors – mean GPA is 2.5Is the difference we observed between the two groups dependable, or did we observe it by chance?Is there really a difference in the underlying population?
Hypothesis Testing
Null hypothesis: There is no difference between these two groupsP-value– the probability of obtaining a difference as big as the one observed when there actually is no differenceWe typically set cutoff at .05What does this mean?
What are Basic Statistical Tests that Developmental Psychopathologists Use?
(1) Differences between groupsT-tests (two groups)ANOVA (more than two groups)
(2) Are continuous variables related to each other?Are greater number of conduct symptoms associated with poorer language skills?Is age associated with depressive symptoms?Relation between two continuous variables: correlation (r)A correlate (book definition): variables associated at a particular point in time No clear proof that one precedes the other – aka no causality*Null hypothesis: r = 0
see plots
What if I Have More than Two Variables
Do conduct disorder symptoms in childhood and SES predict educational attainment?Outcome/Dependent/Y Variable: Number of years of schoolPredictors/Independent/X Variables: Conduct disorder symptoms, family SESMultiple RegressionMultiple IVs and one DV*Unique association of each IV with the DV (β)
Multiple regression
see image
Risk & Protective Factors
Risk factors:Increase the chance of a negative outcome Protective factors: Decrease the chance of a negative outcome
However, there are different types of protective factorsA = protectiveB = protective-stabilizingC = Protective-enhancingD = Protective-reactive
see images
Main Effects & Interactions
Often we have more than one independent variableConduct disorder and socioeconomic statusMain effectAssociation between independent variable and dependent variableConduct disorder and educational attainmentFamily income and educational attainment*Interaction/ModerationThe association between one IV and the DV varies as a function of the other variable
Interactions/ModerationThe association between an IV and the DV depends on the value of the other IVthe association between conduct disorder and educational attainment varies as a function of SESDepressionDo adolescents experience a major depressive episode over the next 36 months?GenderPubertal statusGender x Pubertal Status
Mediator variables
*impact the process, mechanism, or means through which a variable produces a particular outcomeAccount for some (partial mediation) or all (full mediation) of the apparent relationship between two variablesConceptual ViewMediator variables
Summary
We work with samples that we think (or hope) is drawn from a population of interest(2)Inferential statistics tell us whether what we observe in our sample is representative of the underlying population(3)Anytime you read a study:(a)Identify your IVs and your DVs(b)Identify the main effects, interactions, and if there are any mediators Think of stats like detective work! = a tool to help you uncover an answer to a question you care abou
