Assignment 4

Correlations and Regression✓
McLayne Bennett ✓
ID: T00707610
Assignment No. 4✓
Submitted to
Dr. John Marasigan
In partial fulfillment of the requirements for
Psychology 2101
TRU–Open Learning
July 21, 2023
Question 1 Mark: 15.5/18
A) A correlation is used to look at evidence of a relationship between

two variables. Its characteristics include direction, which can either
be positive (when x variable increases then the y variable increases
equally) or negative (when x variable increases then the y variable
decreases). Correlation has linear form in that you can draw a line
through the middle of the scatter plot and the line is either moving up
or moving down; this will demonstrate whether there is a negative or
positive relationship between the variables. Lastly, correlations will
show some sort of pattern and predictability* throughout the scatter
plot, if there is predictability the value of r will be closer to 1 or -1. If
there is no predictability then the correlation value will be close to
zero which would show random points with no linear connection on
the plot. ✓+3.5/4
B) Prediction: Correlation could be used to measure adverse childhood

experiences and crime rates. Based on the amount of ACE’s in
certain populations the correlation could be used to predict potential
criminal activity in communities.
Validity: Measuring effectiveness of psychedelic therapy session by

comparing outcome scores with already established treatment
modalities that were of the same frequency/duration.
Reliability: Measuring reaction times of adults receiving doses of

Adderall months apart and receiving the same or similar scores.
Looking to see if the treatment of ADHD medications increased
reaction times reliably.
Theory Verification: Stating that excessive use of Tik-Tok among 13

year old females increases likelihood for self-harm and then
conducting an experiment with an appropriate sample to investigate
this theory and to see if high Tik-Tok usage is in fact related to self-
harm in 13 year old girls. ✓6/6
C) The Pearson Correlation measures the variability of two groups

together and apart and uses these values in a formula to determine how
closely the variables affect each other. It is used to see if there is a pattern
and if the variables affect each other, it measures the linear relationship of
the variables on a scatter plot; those that form a perfect upward or
downward line can be considered to have a score of 1 which indicates high
correlation. An example would be a score of 1 using the Pearson
Correlation which measured the relationship between screen time and self-
harm in a sample of 10 teenage girls aged 13 years old.
On the other hand, the Spearman Correlation measures the relationship of

two variables that are either measured as ranks or have poor linear form
but there is suspicion that the variables are related and move together in
one direction. Spearman Correlation measures consistency rather than the
strength of relationship and by using ranks or converting scores to ranks,
we can find if the variables do indeed affect each other in a patterned
manner. The Spearman Correlation may be used to look at the relationship
between athletes who place 1st, 2nd, 3rd, 4th, or 5th, and their self-rated
happiness (a score of 1 to 5). ✓4/4
D) An example of a linear relationship between X and Y would be daily Tik

Tok use and GAD-7 scores. The Tik Tok use variable would be the X
variable and measured in full hours. X would only equal 1 if the individual
used Tik Tok for a complete 60 minutes. The Y variable would be the GAD-
7 scores and for this sample there was a selection of individuals with a
GAD score of 5 as the minimum and this individual was someone who did
not use Tik Tok for a whole hourY.* ✓+2/4
Question 2 Mark: 15/15
A) SP= ∑XY - ∑X∑Y/n
X Y XY
7 83 581
11 92 1012
8 87 696
8 78 624
6 74 444
5 69 345
10 85 850
13 93 1209
4 68 272
9 81 729
∑X= 81 ∑Y= 810 ∑XY= 6762
SP = 6762 - 81(810) / 10 = 6762 - 6561 = 201

SP = 201
SS= ∑X^2 - (∑X)^2/10
49+121+64+64+36+25+100+169+16+81 = 725
81^2 /10 = 656.10
SSx= 725 - 656.10 = 68.90
6889+8464+7569+6084+5476+4761+7225+8649+4624+6561 = 66302
810^2/10 = 65610
SSy= 66302 - 65610= 692
r= 201 / √68.90 (692)
201/218.355 = 0.92
r= 0.92 ✓13/13
B) For a two tailed test with ɑ .05 the critical value for df= 8 would be .632,
therefore the correlation of r= .92 between number of hours of study and
test scores could be considered significant. ✓2/2
A)
No Hrs Test Ranked Ranked X^2 Y^2 X(Y)

Study Score No. Hrs Test
(X) (Y) Study Score
(X) (Y)
7 83 4 6 16 36 24
11 92 9 9 81 81 81
8 87 5.5 8 30.25 64 44
8 78 5.5 4 30.25 16 22
6 74 3 3 9 9 9
5 69 2 2 4 4 4
10 85 8 7 64 49 56
13 93 10 10 100 100 100
4 68 1 1 1 1 1
9 81 7 5 49 25 35
∑X= 55 ∑Y=55 ∑X^2= ∑Y^2= ∑XY=

384.5 385 376
SSx= 384.5 - 302.5 = 82
SSy= 385 - 302.5 = 82.5
SP= 376 - 302.5 = 73.5
rs= 73.5/ √ 82(82.5)=73.5 /82.25=0.89 ✓ 13/13
B)
For a two-tailed test with an alpha level of .05, the critical score for df=8
is .632. Since our Spearman calculation = 0.89 we can reject the null
hypothesis and state that this correlation between number of hours of study
and test scores is significant. This calculation is similar to the Pearson
formula used above and most likely a bit smaller as we did not use exact
scores. ✓2/2
Question 4 Mark: 14.5/20
A)
Runners Scores (X) Group (Y) X(Y)

Coffee= 1
Non-coffee = 0
A 7.5 1 7.5
B 8 1 8
C 8.3 1 8.3
D 7.2 1 7.2
E 9.6 1 9.6
F 7.8 0 0
G 7.9 0 0
H 8 0 0
I 7.1 0 0
J 9.4 0 0
∑X= 80.80 ∑Y= 5 ∑XY= 40.60

∑X^2= 659.16 ∑Y^2= 25
SSx= 659.16 - 652.864 = 6.296
SSy= 5 - 2.5 = 2.5
SP= 40.60 - 80.8 (5)/10 = 0.2
r= 0.2/ √6.296 (2.5) = 0.2 / 3.967 = 0.05 ✓13/13
B)
r^2= 0.0025
The variability on the running speeds which is caused by coffee drinking is

incredibly small(.25%) and does not appear significant. ✓+1.5/2
A)
Degree (X) Promotion (Y) X(Y)
0 1 0
0 1 0
0 0 0
1 1 1
1 1 1
1 1 1
1 1 1
1 1 1
1 1 1
1 1 1
1 1 1
0 0 0
0 0 0
1 1 1
0 0 0
1 0 0
1 1 1
1 1 1
1 1 1
0 0 0
∑X= 13 ∑Y= 14 ∑X(Y)= 12
SSx= 13 - 169/20 = 4.55
SSy= 14 - 196/20 = 4.2
SP= 12 - 13(14)/20 = 2.9
r= 2.9 / √13(14) = 2.9 / 13.491 = 0.215 +6/13 (effort)

2.9
= (4.55)(4.2)
√
2.9
= 4.371 =0.663
B)
r^2= 0.464
The r^2 of 0.464 indicated that 46.40% of the variance of whether an

individual received a promotion can be related due to whether they had a
degree or not. Looking at this, it does appear significant but not of major
significance. +1/2 (effort)
r2 = (0.663)2 = 0.44
There is a medium correlation between having a bachelor’s degree and being promoted.
The coefficient of determination indicates that the variability of being promoted as
determined by its relationship with having a bachelor’s degree is 44%.
X - years Y - annual X^2 Y^2 XY

employed salary
36 2500 300
6 50
49 3600 420
7 60
16 1600 160
4 40
9 1225 105
3 35
25 2500 250
5 50
1 900 30
1 30
64 4225 520
8 65
36 2025 270
6 45
4 1225 70
2 35
81 4900 630
9 70
∑X= 51 ∑Y= 480 ∑X^2= 321 ∑Y^2= ∑XY= 2755

Mx= 5.1 My = 48 24700
A)
b= SP/SSx
SSx = 321 - 51^2/10 = 321 - 260.1 = 60. 9
SP = 2755 - 51 (480) /10 = 2755 - 2448 = 307
b= 307/60.9 = 5.04 ✓
a= My=Mx(b)
a= 48 - 5.1 (5.04) = 48 - 25.704 = 22.296 or 22.3✓

ŷ= 5.04X + 22.3 ✓10/10
B) ŷ= 5.04 (5) + 22.3 = 47.5

Based on above regression equation, the predicted annual salary amount
for an individual employed for 5 years is 47,500 dollars. ✓2/2
C) F = MSregression/MSresidual
SSregression = r^2(SSy)
SSy = 24700 - 480^2/10 = 24700 - 23040 = 1660
r= 307 / √60. 9 (1660) = 307/312.953 = 0.981

r^2= 0.9624
SSregression = 0.9624(1660)= 1597.584
SSresidual = (1 - 0.9624) 1660 = 62.416
MSregression = 1597.584/1 = 1597.584
MSresidual = 62.416 / 10-2 = 7.802
F = 1597.584/7.802 = 204.766 +4/9 (effort)
r=SP/√SSXSSY
= 307/ √ (60.9)(1660)
= 307/317.953
= 0.966
SSregression=r2SSY = (0.966)2 (1660)
= 0.933 (1660)
= 1548.78
SSresidual = (1 – r2)(SSy)
SSresidual = (1 – 0.9662)(1660)
= (1 – 0.933) (1660)
= 111.22
MSregression=SSregression/dfregression
= 1548.78/1
=1548.78
MSresidual=SSresidual /dfresidual
= 111.22/8
= 13.9
F=MSregression/MSresidual
= 1548.78/13.9
= 111.423
With df (1, 8) the critical F value for an alpha level of .05 = 4.84 and the F
value for our regression equation is 204.766. It can be concluded that yes,
the regression equation accounts for a major portion of the variance in the
annual salary. ✓ Additionally, if we look at the r^2 of 96.24% we can also
conclude that a significant portion of the variance is due to the effect that
years employed has on an employee’s salary.
TOTAL mark calculation: 84/100 = 84%

Assignment 4

Uploaded by

Copyright:

Available Formats

Assignment 4

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Assignment 4

Uploaded by

Copyright:

Available Formats

Correlations and Regression✓

A) A correlation is used to look at evidence of a relationship between

B) Prediction: Correlation could be used to measure adverse childhood

Validity: Measuring effectiveness of psychedelic therapy session by

Reliability: Measuring reaction times of adults receiving doses of

Theory Verification: Stating that excessive use of Tik-Tok among 13

C) The Pearson Correlation measures the variability of two groups

On the other hand, the Spearman Correlation measures the relationship of

D) An example of a linear relationship between X and Y would be daily Tik

A) SP= ∑XY - ∑X∑Y/n

∑X= 81 ∑Y= 810 ∑XY= 6762

SP = 6762 - 81(810) / 10 = 6762 - 6561 = 201

SS= ∑X^2 - (∑X)^2/10

Question 3 Mark: 15/15

No Hrs Test Ranked Ranked X^2 Y^2 X(Y)

∑X= 55 ∑Y=55 ∑X^2= ∑Y^2= ∑XY=

SSy= 385 - 302.5 = 82.5

SP= 376 - 302.5 = 73.5

rs= 73.5/ √ 82(82.5)=73.5 /82.25=0.89 ✓ 13/13

Question 4 Mark: 14.5/20

Runners Scores (X) Group (Y) X(Y)

∑X= 80.80 ∑Y= 5 ∑XY= 40.60

SSx= 659.16 - 652.864 = 6.296

SSy= 5 - 2.5 = 2.5

SP= 40.60 - 80.8 (5)/10 = 0.2

r= 0.2/ √6.296 (2.5) = 0.2 / 3.967 = 0.05 ✓13/13

The variability on the running speeds which is caused by coffee drinking is

SSy= 14 - 196/20 = 4.2

SP= 12 - 13(14)/20 = 2.9

r= 2.9 / √13(14) = 2.9 / 13.491 = 0.215 +6/13 (effort)

The r^2 of 0.464 indicated that 46.40% of the variance of whether an

X - years Y - annual X^2 Y^2 XY

∑X= 51 ∑Y= 480 ∑X^2= 321 ∑Y^2= ∑XY= 2755

SSx = 321 - 51^2/10 = 321 - 260.1 = 60. 9

SP = 2755 - 51 (480) /10 = 2755 - 2448 = 307

a= 48 - 5.1 (5.04) = 48 - 25.704 = 22.296 or 22.3✓

B) ŷ= 5.04 (5) + 22.3 = 47.5

SSy = 24700 - 480^2/10 = 24700 - 23040 = 1660

r= 307 / √60. 9 (1660) = 307/312.953 = 0.981

SSregression = 0.9624(1660)= 1597.584

SSresidual = (1 - 0.9624) 1660 = 62.416

MSregression = 1597.584/1 = 1597.584

MSresidual = 62.416 / 10-2 = 7.802

F = 1597.584/7.802 = 204.766 +4/9 (effort)

TOTAL mark calculation: 84/100 = 84%

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.