Active Learning Task 8 (ALT 8): Statistical Inference

100 Points, Multiple Choice (4 points each); item 10 (14 points) items 11 & 12 (25 points, each)

Below are some formulae you may or may not need for ALT 8:

σ
σ X=
√n
s
sX =
√n
X −μ
ZX=
σX

X −μ
ZX=
sX

X −1.96 ( s X ) ≤ μ ≤ X + 1.96 ( s X )

X −2.58 ( s X ) ≤ μ ≤ X + 2.58 ( s X )

Number of possible occurrences of target event A .

Probability (of A) =
Total number of possible event outcomes .

Number of occurrences of the score .

Relative Frequency (of a score) =
Total number of possible scores .

X - µ
Z obs =
s

X −X
Z obs=
s

Range = XHighest – XLowest , IQR = X75 – X25

ΣX ΣX
SIQR = IQR/2 = (X75 – X25)/2, μ= , X=
N n

∑ ( X−μ ) 2 ∑ ( X−μ ) 2
σ 2
=
N
σ=
√ N
 ∑ ( X− X ) 2 ∑ ( X−X ) 2
s2 ¿
n−1
s¿
√ n−1

Identify your answer to each item by bolding the alternative of that item:

1. Which of the following is true of the Central Limits Theorem?

a. it promises that the sampling distribution of the mean will be about normal in shape
b. it promises the µ of the sampling distribution will be equal to µ of the population
c. it promises the of the sampling distribution will be less than the of the population by an
amount equal to the square root of the sample size.
d. all of the above

2. The standard deviation of a distribution of sample means is:

a. equal to the population standard deviation divided by the square root of N
b. called the standard error of the mean
c. decreases in size as N increases
d. all of the above

3. What happens to the variance of a sampling distribution, as we decrease the number of people that
each sample is based on?
a. it stays the same as the spread of the original set of scores
b. it becomes more spread out
c. it becomes less spread out
d. the variance of a sampling distribution is not affected by N.

4. Simple random selection is:

a. where each member of the population has an equal chance to be in the research
b. where the selection of one person is independent of the selection of every other person
c. where the selection of one person is affected by the selection of other persons
d. both a and b are true
e. both a and c are true

5. Stratified random selection is:

a. where each member of the population has an equal chance to be in the research
b. where the selection of one person is independent of the selection of every other person
c. where the selection of one person is affected by the selection of other persons
d. both a and b are true
e. both a and c are true

6. The use of sample statistics to draw conclusions about population parameters is called?
a. Descriptive statistics.
b. Hypothesis testing.
c. A sampling distribution.
d. Statistical inference.
7. An estimator that is more likely to be close to the population parameter as the number of
participants in a sample increase, is a?
a. Unbiased estimator.
b. Biased estimator.
c. Consistent estimator.
d. Inconsistent estimator.

8. If an instructor plans to combine test grades from three exams, and assign course grades on
a curve (only 10% A, 20% Bs, etc.) he should first?
a. Test to see if each exam is normally distributed.
b. Convert all scores to percentile ranks.
c. Convert all scores to deviation scores.
d. Convert all scores to standard scores or z-scores

9. The reason why it is preferable to covert raw scores to z-scores when combining results from
separate measures into a single composite is?
a. z-scores are interval scales of measurement.
b. z-scores produce symmetrical distributions with no skew.
c. all z-score distributions have a mean of zero.
d. all z-score distributions have a standard deviation of one.
e. c and d.

A researcher interested in cognitive aging decides to test the hypothesis that slowing of mental
processes with age is due to loss of aerobic capacity, impairing blood flow and nutriment supply
to the brain. He recruits 20, 70-year-old adults from the population of Americans to participate
in a test of his research hypothesis.

He provides each of the 20 participant with a stationary bicycle for home use and engages them
in a monitored exercise routine for a 6-month period. The first two weeks they cycle at low
intensity for 15-minutes, 5 times a week. After two weeks, they increase to 5, 30-minute periods
of a moderate intensity level of cycling each week. Beginning with the second month, they cycle
at moderate intensity for 60-minutes a day, 5 times a week. This continues for the next five
months.

At the end of 6 months he measures these 20 participants on visual processing speed using a
backward masking task (to millisecond precision or 1/1000th of a second).

He knows from previous research that the μ visual processing time of 70-year-old adults (who do
not exercise) is 277 milliseconds with a σ equal to 25.0 milliseconds.

The data he collected from the 20, 70-year-old exercisers is shown in the table below:
Visual Information Processing Data for 20, 70-year-olds who did extensive aerobic exercise
training for six months. The numbers in the table have also been copied to the Excel Template
for Active Learning Task 8

Row Exercise
Labels X ( X −X ) ( X −X ) 2
1 300.00 37.00 1369.00
2 295.00 32.00 1024.00
3 290.00 27.00 729.00
4 285.00 22.00 484.00
5 280.00 17.00 289.00
6 275.00 12.00 144.00
7 270.00 7.00 49.00
8 270.00 7.00 49.00
9 265.00 2.00 4.00
10 265.00 2.00 4.00
11 265.00 2.00 4.00
12 260.00 -3.00 9.00
13 260.00 -3.00 9.00
14 255.00 -8.00 64.00
15 250.00 -13.00 169.00
16 245.00 -18.00 324.00
17 240.00 -23.00 529.00
18 235.00 -28.00 784.00
19 230.00 -33.00 1089.00
20 225.00 -38.00 1444.00

Sums 5260.0
0 8570.00
Mean
Variance 451.05
Std Dev 21.2379

10. Compute and fill in the missing values for the Mean, Variance and Standard Deviation in
the table above. Copy the formulae you used for these computations, here:
Σ X 2 ∑ ( X− X ) 2 ( X−X ) 2
s¿ ∑
X=
n
s ¿
n−1 √ n−1

4.4721
 11. This researcher wants to decide if six months of aerobic exercise training resulted in a
decrease in the visual information processing times of his sample of 20, 70-year-old
participants. He thinks this question could be answered by knowing the probability that
a mean performance equal to or smaller than his sample ( X ), could have been observed
in the population of 70-year-olds who do not exercise (μ = 277, σ =25.0).
(1) Label the relevant values in the sketch below,
(2) Compute this probability,
(3) show all of the formulae involved in computing this probability and
(4) explain why you think these data support or refute his research hypothesis. You can
use the “insert” “textbox” options in WORD to label the sketch values below.
X −μ s
ZX= = -2.948 = z score 0.0016 s X = = The 4.749
sX √n data supports his hypothesis because
the probability of getting a mean
performance equal to or smaller than his
sample has a 0.0016 probability.

25.0

277
263
 12. The researcher thinks knowing the 95% confidence interval for the sample mean ( X ) of
his 20, 70-year-old participants would help him decide if aerobic exercise decreased the
visual information processing times of his sample.
(1) Show all the formula required for this computation,
(2) Compute the 95% confidence interval using the data from problems 10 & 11 above
(3) relate the bounds of the confidence interval you compute to the known mean ( μ ) of
the population, which is 277 milliseconds, as listed above. Do you think his data
supports the research hypothesis that aerobic exercise decrease processing times?

X −1.96 ( s X ) ≤ μ ≤ X + 1.96 ( s X )
263−1.96 ( 4.749 ) ≤ μ ≤ 263+1.96 ( 4.749 )
263-9.308 263+9.308
253.692≤ μ ≤ 272.308

We can be 95% confident that the mean reaction time of 70-year-olds who exercised is
not below 253.692 and not above 272.308.
This supports the hypothesis because the mean for the non-exercising group is outside the
boundary.