LESSON 17: T-TEST FOR TWO-INDEPENDENT SAMPLES Estimated Standard Error

Research designs that are used to obtain the two sets of data  represented by the symbol S(M -M )
1 2

can be classified in two general categories:

Interpreting the Estimated Standard Error
 two sets of data could come from two completely
Can be interpreted in 2 ways:
separate groups of participants
o ex. study that involves a sample of men  a measure of the standard or average distance
compared with a sample of women between a sample statistic (M1 − M2) and the
 two sets of data could come from the same group of corresponding population parameter (μ1 − μ2).
participants  measures the standard, or average size of (M 1 − M2)
o ex. one set of scores measuring depression if the null hypothesis is true. That is, it measures how
for a sample of patients before they begin much difference is reasonable to expect between the
therapy and then obtain a second set of data two-sample means.
by measuring the same individuals after 6 III. Calculating the Estimated Standard Error
weeks of therapy
Three points:
INDEPENDENT-MEASURES RESEARCH DESIGN
(BETWEEN-SUBJECTS DESIGN)  Each of the two-sample means represents its own
population mean, but in each case there is some error
 research design that uses a separate group of o M1 approximates μ1 with some error
participants for each treatment condition (or for each o M2 approximates μ2 with some error
population)  Thus, there are two sources of
Repeated-Measures Research Design (Within-Subjects error.
Design)  estimated standard error for each sample mean is
computed as follows:
 two sets of data are obtained from the same group of
participants
I. Hypotheses for an Independent-Measures Test  resulting formula for standard error is:
 goal of an independent-measures research study: to
evaluate the mean difference between two
populations (or between two treatment conditions)
 null hypothesis: there is no change, no effect, or, in o limited to situations in which the two samples
this case, no difference. are the same size (that is, n1 = n2)
o H0: μ1 − μ2 = 0 (No difference between the IV. Pooled Variance
population means)  variance obtained from a large sample is a more
 could also be stated as μ1 = μ2 accurate estimate of σ2 than the variance obtained
 alternative hypothesis: there is a mean difference from a small sample.
between the two populations.  Pooled variance – one method for correcting the bias
o H1: μ1 − μ2 ≠ 0 (There is a mean difference.) in the standard error is to combine the two sample
 two population means are not variances into a single value.
equal: μ1 ≠ μ2 o Obtained by averaging or “pooling” the two
II. Formulas for an Independent-Measures Hypothesis Test sample variances using a procedure that
allows the bigger sample to carry more
Two points to remember: weight in determining the final value.
 The basic structure of the t statistic is the same for
both the independent-measures and the single-
sample hypothesis tests
V. Estimated Standard Error
 Resulting formula for ESE
 The independent-measures t is basically a two-
sample t that doubles all the elements of the single-
sample t formulas.  measures how accurately the difference between two
sample means represents the difference between the
two population means.
 value for μ1 − μ2 comes from the null hypothesis: 0  how much difference is expected on average between
the two sample means.
VI.Final Formula II. Hartley’s F-Max Test
 complete formula for the independent-measures t  Used to check homogeneity of variance
statistic:

 df

HYPOTHESIS TESTS WITH HE INDEPENDENT-MEASURES

T STATISTIC

EFFECT SIZE AND CONFIDENCE INTERVALS FOR THE

INDEPENDENT-MEASURES T

 Cohen’s d – produces a standardized measure of

mean difference

 formula for estimating Cohen’s d becomes:

 r2 measures how much of the variability in the scores

can be explained by the treatment effects.
I. Assumptions Underlying the Independent-Measures t
Formula
 The observations within each sample must be LESSON 18: T TEST FOR TWO RELATED SAMPLES PT1
independent INTRODUCTION TO REPEATED-MEASURES DESIGN
 The two populations from which the samples are
selected must be normal.  Repeated-measures designs (within-subject design) –
 The two populations from which the samples are one in which the dependent variable is measured two
selected must have equal variances. or more times for each individual in a single sample.
 The same group of subjects is used in all the
treatment conditions.
I. Matched-Subjects Design
 each individual in one sample is matched with an
individual in the other sample.
 matching is done so that the two individuals are
equivalent (or nearly equivalent) with respect to a
specific variable that the researcher would like to
control.
THE T STATISTIC FOR A REPEATED-MEASURES
RESEARCH DESIGN
I. Difference Scores: Data for a Repeated-Measures Study
 difference scores are obtained by subtracting the first
score (before treatment) from the second score (after
treatment) for each person:

II. Hypotheses for a Related-Samples Test I. Assumptions of the Related-Samples t Test

 repeated-measures study: null hypothesis: the mean  The observations within each treatment condition
difference for the general population is zero must be independent. Notice that the assumption of
independence refers to the scores within each
treatment. Inside each treatment, the scores are
obtained from different individuals and should be
 alternative hypothesis: there is a treatment effect that independent of one another.
causes the scores in one treatment condition to be  The population distribution of difference scores (D
systematically higher (or lower) than the scores in the values) must be normal.
other condition.
EFFECT SIZE AND CONFIDENCE INTERVALS FOR THE
REEATED-MEASURES T

III. t Statistic for Related Samples  Estimated d

 T stats

 Percentage of variance
 First, Variance or SD

 Then, ESE  Confidence intervals for estimating 𝛍D

COMPARING REPEATED- AND INDEPENDENT-MEASURES

 Df in t stats: n-1 DESIGNS
LESSON 19: T TEST FOR TWO-RELATED SAMPLES PT 2  Number of subjects:
HYPOTHESIS TEST FOR THE REPEATED-MEASURES o repeated-measures design typically requires
DESIGN fewer subjects
o repeated-measures design uses the subjects
 Ex. more efficiently because each individual is
measured in both of the treatment conditions
 study changes over time
o RMD is well suited for studying learning, or
other changes that take place over time
  individual differences I. Terminology in ANOVA
o primary advantage of a repeated-measures  Quasi-independent variable – non-manipulated
design is that it reduces or eliminates variable
problems caused by individual differences  Factor – the variable that designated the groups
o individual differences: age, IQ, gender, being compared
personality  Levels – individual groups or treatment conditions that
I. Time-Related Factors and Order Effects are used to make up a factor.
 primary disadvantage of a repeated-measures  ANOVA can be used with either an independent-
design: structure of the design allows for factors other measures or a repeated-measures design.
than the treatment effect to cause a participant’s  ANOVA can be used to evaluate the results from a
score to change from one treatment to the next. research study that involves more than one factor
o outside factors that change over time may  two-factor design or a factorial design - study that
be responsible for changes in the combines two factors
participants’ scores o provides researchers with the flexibility to
 participation in the first treatment influences the develop studies that address scientific
individual’s score in the second treatment. questions that could not be answered by a
o Order effects – changes in scored that are single design using a single factor
caused by participation in n earlier II. Statistical Hypotheses for ANOVA
treatment; can distort the mean differences  Null hypothesis: no effect; the population means are
found in repeated-measures research all the same; there is no treatment effect
studies.
 Counterbalancing – one way to deal with time-related
factors and order effects
o Participants are randomly divided into two  Alternative hypothesis: population means are not all
groups with one group receiving treatment 1 the same; there is a real treatment effect
followed by treatment 2; other group
receiving treatment 2 followed by treatment
1.
o Goal: to distribute any outside effects evenly
over the two treatments
o One alternative: would be that the first two
LESSON 20: ANALYSIS OF VARIANCE (ANOVA) PT 1
populations are identical, but that the third is
 hypothesis-testing procedure that is used to evaluate different.
mean differences between two or more treatments (or o Another: the last two means are the same,
populations). but the first is different.
 Major advantage: can be used to compare two or III. Type I Errors and Multiple-Hypothesis Tests
more treatments.  Often a single experiment requires several hypothesis
o provides researchers with much greater tests to evaluate all the mean differences. However,
flexibility in designing experiments and each test has a risk of a Type I error, and the more
interpreting results. tests you do, the greater the risk.
 Goal: to determine whether the mean differences  Testwise alpha level – risk of a Type I error, or alpha
observed among the samples provide enough level, for an individual hypothesis test.
evidence to conclude that there are mean differences  Experimentwise alpha level – involves several
among the three populations. different hypothesis tests; total probability of a Type I
 Two interpretations: error that is accumulated from all of the individual
o There really are no differences between the tests in the experiment; greater than the value of
populations (or treatments). The observed alpha used for any one of the individual tests.
differences between the sample means are IV. Test Statistic for ANOVA
caused by random, unsystematic factors  we first computed the standard error, which measures
(sampling error) that differentiate one the difference between two sample means that is
sample from another. reasonable to expect if there is no treatment effect
o The populations (or treatments) really do (that is, if H0 is true). Then we computed the t statistic
have different means, and these population with the following structure:
mean differences are responsible for
causing systematic differences between the
sample means.
  we can use variance to measure sample mean
differences when there are two or more samples. The
test statistic for ANOVA uses this fact to compute an  value obtained for the F-ratio helps determine
F-ratio with the following structure: whether any treatment effects exist. Consider the
following two possibilities:
o When there are no systematic treatment
effects, the differences between treatments
(numerator) are entirely caused by random,
LOGIC OF ANOVA unsystematic factors; numerator and
denominator are both measuring ransom
differences and should be roughly the same
size.
 Roughly equal numerator and
denominator: it should have a value
around 1.00

 goal: measure the amount of variability and to explain  F-ratio near 1.00: the differences
why the scores are different between treatments (numerator)
 first step: determine total variability for the entire set are random and systematic, same
of data. sa denominator; there is no
o Combine all scores evidence to suggest that the
o Break it apart into separate components treatment has any effect.
 Analysis process divides the total variability into two o When the treatment does have an effect,
basic components: causing systematic differences between
o Between-Treatments Variance – We will samples, then the combination of systematic
and random differences in the numerator
calculate the variance between treatments to
should be larger than the random differences
provide a measure of the overall differences
alone in the denominator.
between treatment conditions; the variance
between treatments is really measuring the  Error term – the denominator of the F-ratio measures
differences between sample means. only random and unsystematic variability
o Within-Treatment Variance – variability within ANOVA NOTATION AND FORMULAS
each sample; provides a measure of the
variability inside each treatment condition.  k – used to identify the number of treatment
I. Between-Treatments Variance conditions – number of levels of the factor.
 Measures how much difference exists between the o For IMS, it is the number of separate
treatment conditions. samples.
 Two possible explanations:  n – number of scores in each treatment
o The differences between treatments are not  N – total number of scores in the entire study
caused by any treatment effect but are  T – sum of the scores ( ΣX) for each treatment
simply the naturally occurring, random and condition
unsystematic differences that exist between  G – sum of all the scores in the research study (grand
one sample and another. That is, the total) G = ΣT
differences are the result of sampling error. I. ANOVA Formulas
o The differences between treatments have  Final calculation for ANOVA is the F-ratio (composed
been caused by the treatment effects. of 2 variances)
II. Within-Treatments Variance
 provides a measure of how big the differences are
when H0 is true.
III. F-Ratio: Test Statistic for ANOVA
 Comparison (between and within treatments) is made  Each of the two variances in the F-ratio is calculated
by computing an F-ratio. using the basic formula for sample variance.
 For independent measures ANOVA:
  the entire process of ANOVA requires nine
calculations: three values for SS, three values for df,
two variances (between and within), and a final F-  MSwithin
ratio
II. Analysis of Sum of Squares (SS)
 Total Sum of Squares (SStotal) – the sum of squares
for the entire set of N scores  F-ratio

LESSON 21: ANOVA PT 2

I. Distribution of F-Ratios
 F-ratio is constructed so that the numerator and
denominator of the ratio are measuring exactly the
same variance when the null hypothesis is true.
o Computational formula: o If the null hypothesis is false, the F-ratio
should be much greater than 1.00
 two obvious characteristics:
o Because F-ratios are computed from two
 Within-Treatments Sum of Squares (SSwithin) – the variances (the numerator and denominator
variability inside each of the treatment conditions of the ratio), F values always are positive
numbers. Remember that variance is always
positive.
o When H0 is true, the numerator and
denominator of the F-ratio are measuring the
 Between-Treatments Sum of Squares (SSbetween)
same variance. In this case, the two sample
variances should be about the same size, so
the ratio should be near 1. In other words,
the distribution of F-ratios should pile up
III. Analysis of DF around 1.00.
 two important considerations to keep in mind: II. F Distribution Table
o Each df value is associated with a specific  For ANOVA, we expect F near 1.00 if H0 is true.
SS value.  F-ratio that is much larger than 1.00 is an indication
o Normally, the value of df is obtained by that H0 is not true.
counting the number of items that were used
POST HOC TEST
to calculate SS and then subtracting 1.
 Total Degrees of Freedom (dftotal)  additional hypothesis tests that are done after an
ANOVA to determine exactly which mean differences
are significant and which are not.
 These tests are done after ANOVA when:
 Within-Treatments Degrees of Freedom (dfwithin) o You reject H0 and
o there are three or more treatments (k ≥ 3)
 Rejecting H0 indicates that at least one difference
exists among the treatments.
 Between-Treatments Degrees of Freedom (dfbetween) I. Posttests and Type I Errors
 Pairwise comparisons – compare the individual
treatments two at a time.
o Ex. with k = 3, we would compare μ 1 vs. μ2,
 Total – entire set of scores
then μ2 vs. μ3, and then μ1 vs. μ3.
 Within – differences that exist inside the individual
 Experimentwise alpha level – as you do more and
treatment conditions.
more separate tests, the risk of a Type I error
 Between – differences from one treatment to another.
accumulates
IV. Calculation of Variances (MS) and the F-Ratio
II. Tukey’s Honestly Significant Difference (HSD) Test
 MSbetween
  allows you to compute a single value that determines o The distribution of t and the distribution of F-
the minimum difference between treatment means ratios match perfectly if you take into
that is necessary for significance. consideration the relationship F = t2.
 HSD – used to compare any two treatment
LESSON 22: OVERVIEW OF CHI-SQUARE STATISTIC
conditions.
 If the mean difference exceeds Tukey’s HSD, you I. Parametric and Nonparametric Statistical Tests
conclude that there is a significant difference between  Parametric tests – tests all concern parameters and
the treatments. require assumptions about parameters
 Otherwise, you cannot conclude that the treatments o they require a numerical score for each
are significantly different. individual in the sample.
 Formula: o require data from an interval or a ratio scale
o require numerical scores
 nonparametric tests – alternatives
 Yung q, galing sa F Dist table o use sample data to evaluate hypotheses
III. Scheffè Test about the proportions or relationships that
 uses an F-ratio to evaluate the significance of the exist within populations.
difference between any two treatment conditions. o do not state hypotheses in terms of a
 “safety factor” for the Scheffé test comes from the specific parameter and they make few (if
following two considerations: any) assumptions about the population
o Although you are comparing only two distribution.
treatments, the Scheffé test uses the value o Distribution-free tests
of k from the original experiment to compute o participants are usually just classified into
df between treatments. Thus, df for the categories
numerator of the F-ratio is k – 1. o involve measurement on nominal or ordinal
o The critical value for the Scheffé F-ratio is scales
the same as was used to evaluate the F- o simply frequencies
ratio from the overall ANOVA. Thus, Scheffé  situations for which transforming scores into
requires that every posttest satisfy the same categories might be a better choice.
criterion that was used for the complete o it is simpler to obtain category
ANOVA. measurements
IV. Relationship Between ANOVA and t Tests o original scores may violate some of the
 relationship between pooled variance and MS within is basic assumptions that underlie certain
only part of the general relationship between the statistical procedures
independent-measures t test and the corresponding o original scores may have unusually high
ANOVA.
variance
 basic relationship between t statistics and F-ratios o an experiment produces an undetermined,
can be stated in an equation:
or infinite, score
II. Chi-Square Test for Goodness of Fit
o t statistic compares distances: the distance  uses sample data to test hypotheses about the shape
between two sample means (numerator) or proportions of a population distribution
and the distance computed for the standard  determines how well the obtained sample proportions
error (denominator). fit the population proportions specified by the null
o F-ratio compares variances. You should hypothesis.
recall that variance is a measure of squared III. Null Hypothesis for the Goodness-of-Fit Test
distance. Hence, the relationship: F = t2.  specifies the proportion (or percentage) of the
 Several other points to consider in comparing the t population in each category.
statistic to the F-ratio:
o you will be testing the same hypotheses
whether you choose a t test or an ANOVA.
 H0 falls into one of the following categories:
Hypotheses for either test are:
o No Preference, Equal Proportions – the
population is divided equally among the
categories.
o The degrees of freedom for the t statistic and
the df for the denominator of the F-ratio
(dfwithin) are identical.
 o No difference from a Known Population –
the proportions for one population are not
different from the proportions than are
known to exist for another population.

 alternative hypothesis (H1) simply states that the

population proportions are not equal to the values
specified by the null hypothesis.
IV. Data for the Goodness-of-Fit-Test
 select a sample of n individuals and count how many
are in each category
 observed frequencies (fo)– resulting values; number
of individuals from the sample who are classified in a
particular category.
V. Expected Frequencies
 frequency value that is predicted from the proportions
in the null hypothesis and the sample size (n).
 define an ideal, hypothetical sample distribution that
would be obtained if the sample proportions were in
perfect agreement with the proportions specified 9in
the null hypothesis.

VI.Chi-Square Statistic
 measures how well the data ( fo) fit the hypothesis ( fe).

 steps:
o Find the difference between fo (the data) and fe
(the hypothesis) for each category.
o Square the difference. This ensures that all
values are positive.
o Next, divide the squared difference by fe.
o Finally, sum the values from all the categories.
 numerator measures how much difference there is
between the data and the hypothesis. (first 2 steps)
 final step – add the values to obtain the total discrepancy
between the data and the hypothesis
o large value = data do not fit the hypothesis: reject
H0
 third step: obtained discrepancy between fo and fe is
viewed as relatively large or relatively small depending on
the size of the expected frequency.