QUIZZES CANVAS MRM2
Week 1 – ANOVA Met opmerkingen [HS1]: Whether different groups score
the same on a quantitative outcome
Slide 1 – 6 https://www.scribbr.nl/statistiek/anova/
Q1: We usually reject the null hypothesis, if the p-value is below .05
Ø True
Q2: Consider the following playful analogy to look at the meaning of the p value. You want to
test whether your pen was hidden (H0: it was hidden – HA: It was not hidden). If your test for
this has a p-value of .8 (80%), it is likely that your pen was hidden.
Ø True
Q3: Consider the following playful analogy to look at the meaning of the p value. You want to
test whether your pen was hidden (H0: it was hidden). If your test for this has a p-value of 1%
(.01), what do you conclude?
Ø Your conclusion is that most likely your pen was NOT hidden. In other words, you reject
the null hypothesis.
Slide 7 – 12
Q4: What is a conceptual model
Ø A visual representations of relations between variables
Q5: Categorical variables are used when
Ø We categorize our observations (respondents) into different groups, based on certain
criteria, for example education level
Q6: Quantitative variables are used when
Ø We measure constructs via scales on which, in theory, you could score any value, for
example height or time
Q7: Which is correct?
Ø A positive effect of a PV on an OV, means that when the PV increases (or decreases),
the OV does the same (PV; predictive variable & independent) (OV; Outcome variable
& dependent)
Q8: Which is correct?
Ø A negative effect of a PV on an OV, means that when the PV increases (or decreases),
the OV does the opposite
,Slide 13 – 24
Q9: The idea behind an ANOVA is to statistically investigate
Ø Whether different groups score the same on a quantitative outcome
Q10: If scores on a quantitative outcome vary more WITHIN groups than BETWEEN groups,
for example exam scores for three groups of students who each received a different teaching
method. It then is unlikely that it matters which group you are in, regarding your score on the
outcome
Ø True
Q11: To perform an ANOVA the PV and OV need to be measured in the following ways:
Ø The OV is quantitative (continuous) and PV is categorical, with more than two
categories (levels)
Q12: The sum of squares (SS) calculations are a
Ø Quantification of variability of scores of a quantitative outcome
Q13: The total sum of squares represents
Ø The total variability in scores on an outcome variable, on an outcome variable
Q14: The sum of squares of the model PLUS the sum of squares of the residual, together add
up to
Ø The total sum of squares
Q15: The R squared is a measure that quantifies Met opmerkingen [HS2]: how well the regression model
Ø The proportion of our total variability in the OV that can be explained by the model explains observed data.
Q16: The F statistic is a ratio of
Ø The EXPLAINED variance to the UNEXPLAINED variance. It tells us how much more
explained variance we have compared to unexplained variance
Q17: The null hypotheses of the F test is:
Ø All the mean scores of groups on the outcome variable are the same
Slide 26 – 34
Q18: The Levene's test for statistical significance test whether
Ø The variance in the groups are equal
Q19: Obtaining a p-value smaller than 5% (.05) for the F test in an ANOVA tells us that
Ø We can confidently reject the null hypothesis, meaning that AT LEAST one of the group
mean differs from the others
Q20: Obtaining an F value of 11 tells us that
Ø We have 11 times more EXPLAINED variance than UNEXPLAINED variance in our
dataset
,Q21: Obtaining an R squared of 0.16 tells us that
Ø We can explain 16% of the differences in scores on our OV, by our model
Q22: When conducting a follow up test to the F test, and examining graphical information
through a means plot, we are able to obtain
Ø The different average scores of the groups on the outcome variable and see which
groups score higher/lower than which other groups
Q23: When conducting multiple comparisons, we
Ø Test whether the mean scores of individual groups statistically differ from one another
Q24: The column "mean difference" in the multiple comparisons output gives us information
on
Ø The difference in mean scores of the two groups that are compared, in terms of the
size of the difference and which group scores higher/lower than which other group
Q25: The "95% confidence interval" of a comparison of two means, in the multiple
comparison table, tells us that
Ø We can be 95% confident that we will find a mean difference that falls within this
interval (lower to upper limit) in the population
Week 2 – FACTORIAL ANOVA Met opmerkingen [HS3]: https://www.scribbr.nl/statistiek/
anova/
Slide 1 – 8
Q1: When we speak of an 'effect' or 'impact' of a categorical PV on a quantitative (continuous)
OV, we then look for
Ø Differences in averages scores on the outcome variable between categories of the
predictor variable
Q2: In a multiple comparisons post hoc test, the 95% confidence interval can be interpreted
as a 'bandwidth of certainty' of the single value estimate for the mean difference
Ø True
Q3: When we do a post hoc test for a one-way ANOVA with a PV that has three categories,
how many unique comparisons can we make?
Ø 3
Slide 9 – 15
Q4: Moderation is statistically and mathematically identical to interaction Met opmerkingen [HS4]: Interaction = moderation
Ø True (mathematically they are identical, however conceptually we
may distinguish between the two) • The effect of one PV on
the OV is moderated by another PV. • The effect of one PV
Q5: The idea behind moderation is that the effect of one PV depends on the value of another on the OV depends on the level of another PV. • PV’s
PV interact in their effect on the OV.
Ø True
, Q6: Reflecting on the knowledge you gained, it would NOT make sense to frame moderation
as a so called "conditional effect" Met opmerkingen [HS5]: Another term for interaction
Ø False effect.
Q7: When our model contains interaction, an INTERACTION effect occurs when the direction
AND/OR the size of the effect of one predictor on an outcome changes, depending on the
value of the other predictor
Ø True
Slide 16 – 25
Q8: The 'Factorial' in Factorial ANOVA refers to something we call 'factors', based on your
current knowledge, what would you say this stands for
Ø It is a synonym for predictor variables or independent variables
Q9: To perform a Factorial ANOVA the PV and OV need to be measured in the following ways
Ø The OV is quantitative (continuous) and PV is categorical, with more than two
categories (levels)
Q10: The idea behind the R squared in Factorial ANOVA differs from that of a one way ANOVA
Ø False
Q11: The sum of squares of the model in a Factorial ANOVA can be subdivided into separate
components which represent
Ø The sum of squares of the separate PVs and the interaction
Q12: In a Factorial ANOVA with 2 PVs and interaction, there are a total of four meaningful F
ratios that we can calculate
Ø True
Q13: The null hypotheses of the F test for the main effect of a single PV in a Factorial ANOVA Met opmerkingen [HS6]:
is Met opmerkingen [HS7]: NO
Ø All the mean scores of groups on the outcome variable are the same
Q14: The null hypothesis of the F test for interaction in a Factorial ANOVA is
Ø There is no interaction
Q15: The total sum of squares represents
Ø The variability in scores on an outcome variable, that we can attribute to our model
and the residual
Q16: The sum of squares (SS) calculations are a
Ø Quantification of variability of scores of a quantitative outcome
