REGRESSION

ANALYSIS

Classified as Internal
 Rsgísssio⭲

o is a sľaľisľical ľsc»"iq"s "ssd i" písdicľi"g ľ»s bs»a:ioí of

a :aíiabls.

o is ssss"ľiall;, a CAUSE-AND-EIIECľ íslaľio"s»ip

o is ľ»s píocsss of písdicľi"g :aíiabls Y (dsps"ds"ľ

:aíiabls) "si"g :aíiabls X (i"dsps"ds"ľ :aíiabls)

2
Classified as Internal
 DEPENDENľ (Y) s INDEPENDENľ (X)
VARIABLES
Independent Variable
Dependent Variable (Response
(Predictor)
Variable)

-is a value that is affected -is a value that the

researcher can freely
when the value of the
change to test its effect
independent variable
on the dependent
changes.
variable.
EXAMPLES: INDEPENDENT VARIABLE
DEPENDENT VARIABLE
- Time spent in studying causes a change in Test Score
-
INDEPENDENT DEPENDENT
VARIABLE (Nutrient intake) influences (growth of an infant) VARIABLE

• I" coííslaľio" a"d ísgísssio" a"al;sss, iľ is impoíľa"ľ ľ»aľ i"dsps"ds"ľ a"d dsps"ds"ľ :aíiablss aís
coííscľl; ids"ľifisd ľo maks coííscľ co"cl"sio"s. I"ľsíc»a"gi"g ľ»sss 2 :aíiablss mosľ of ľ»s ľims ís"dsís 3
ľ»s scaľľsí ploľs a"d coííslaľio" cosfficis"ľs msa"i"glsss.
Classified as Internal
 Remember the acronym: DRY-MIX

D-dsps"ds"ľ :aíiabls M-manipulated variable or the

k-ísspo"di"g :aíiabls one that is changed in an
Y-is ľ»s axis o" w»ic» ľ»s experiment
dsps"ds"ľ oí ísspo"di"g I-independent variable
:aíiabls is gíap»sd (ľ»s X-is the axis on which the
:síľical axis independent variable is graphed
(the horizontal axis

Classified as Internal
 4

Classified as Internal
 Answer: Determine the Answer: Determine the
VARIABLES DV & IV
1. study time (a) and grades (b) 1. a. IV, b. DV
2. salary (a) and educational attainment (b) 2. a. DV, b. IV
3. sales (a) and advertising & marketing (b) 3. a. DV, b. IV
4. effect of soda (a) on blood sugar level (b) 4. a. IV, b. DV
5. test score (a) and tutoring (b) 5. a. DV, b. IV
6. investment choices (a) and risk appetite (b) 6. a. DV, b. IV
7. fasting (a) and body weight (b) 7. a. IV, b. DV
8. effect of phone usage (a) before bedtime on 8. a. IV, b. DV
number of hours of sleep (b)
9. employment rate (a) and minimum wage (b) 9. a. DV, b. IV
10. effect of caffeine (a) on sleep (b) 10. a. IV, b. DV
 SCAľľER PLOľS

• ksgísssio" A"al;sis ísq"iíss i"ľsí:al a"d íaľio-ls:sl daľa.

• ľo sss if ;o"í daľa IIľS ľ»s modsls of ísgísssio", iľ is wiss
ľo co"d"cľ a scaľľsí ploľ a"al;sis.

• ksaso": ksgísssio" a"al;sis ass"mss li"saí íslaľio"s»ip. If

;o" »a:s a c"í:ili"saí íslaľio"s»ip oí "o íslaľio"s»ip,
ísgísssio" a"al;sis is of liľľls "ss.

6
Classified as Internal
 ľ;pss or Li⭲ss

7
Classified as Internal
 Rsgísssio⭲ Li⭲s
o is ľ»s bssľ sľíaig»ľ-li"s dsscíipľio" of ľ»s ploľľsd poi"ľs
a"d ca" "ss ľo dsscíibs ľ»s associaľio" bsľwss" ľ»s
:aíiablss. If all ľ»s daľa poi"ľs fall sxacľl; o" ľ»s li"s, ľ»s"
ľ»s li"s is 0 a"d ;o" »a:s a psífscľ íslaľio"s»ip.

REGRESSION
• Calculates the “best-fit” line (regression line) for a certain set
of data.
• The regression line makes the sum of the squares of the
residuals smaller than for any other line
• Regression minimizes residuals
®
Classified as Internal
 Rsgísssio⭲
o ws aís abls ľo co"sľí"cľ a bssľ fiľľi"g sľíaig»ľ li"s ľo ľ»s
scaľľsí diagíam poi"ľs a"d ľ»s" foím"laľs a ísgísssio"
sq"aľio" i" ľ»s foím of:

Where:
y = dependent variable
x = independent variable
b = intercept
a = slope
tı
Classified as Internal
 ľHINGS ľO REMEMBER:

o ksgísssio" sľill foc"sss o" associaľio", "oľ ca"saľio".

o Associaľio" is a "scsssaí; písísq"isiľs foí i"fsííi"g
ca"saľio" b"ľ also:
o ľ»s i"dsps"ds"ľ :aíiabls m"sľ bs píscsdsd b; ľ»s
dsps"ds"ľ :aíiabls i" ľims.
o ľ»s ľwo :aíiablss m"sľ bs pla"sibl; li"sd b; a ľ»soí;
o Compsľi"g i"dsps"ds"ľ :aíiablss m"sľ bs slimi"aľsd

10
Classified as Internal
 Rsgísssio⭲ Cosrricis⭲t

o is ľ»s slops of ľ»s ísgísssio" li"s a"d ľslls ;o" w»aľ "aľ"ís
of ľ»s íslaľio"s»ip bsľwss" ľ»s :aíiabls is.
o »ow m"c» c»a"gs i" ľ»s i"dsps"ds"ľ :aíiablss is
associaľsd wiľ» »ow m"c» c»a"gs i" ľ»s dsps"ds"ľ
:aíiabls.
o ľ»s laígsí ľ»s ísgísssio" cosfficis"ľ, ľ»s moís c»a"gs.
o »ows:sí, ľ»s ísgísssio" cosfficis"ľ is "oľ a good
i"dicaľoí foí ľ»s sľís"gľ» of íslaľio"s»ip bsca"ss ľwo
scaľľsí ploľs wiľ» :sí; diffsís"ľ dispsísio"s co"ld
píod"cs ľ»s sams
11
Classified as Internal
 ísgísssio" li"s.

12
Classified as Internal
 2 BASIC
ÏORMS OÏ
REGRESSION
Classified as Internal
ANALYSIS
 ÏORMS OÏ REGRESSION ANALYSIS
Exampls: Yo" wa"ľ ľo k"ow oí písdicľ w»aľ i"fl"s"css a psíso"’s salaí;.

EDUCATIONAL
ATTAINMENT

PURPOSE:
WORKING  Measurement of the
HOURS SALARY influence of one or more
variables on another
variable.
Dependent  Prediction of a variable by one
AGE Variable (Criterion) or more other variables

Independent Variables
Source: DATAtab. (2021, February 8). Simple and Multiple Linear Regression [Video].
(Predictors) https://www.youtube.com/watch?v=29rjWClT_3U
 LINEAR s MULľIPLE LINEAR REGRESSION
Simple Linear Multiple Linear
 Do the weekly working hours and the age of employees
 Does the weekly working time have an influence on
have an influence on their hourly salary?
the hourly salary of the employees?

LINEAR REGRESSION MULTIPLE LINEAR REGRESSION

- also called simple linear regression. - considers more than one quantitative and qualitative
-considers one quantitative and independent variable (x) to variable to predict a quantitative and dependent variable y.
predict the other quantitative, but dependent, variable y.

Source: DATAtab. (2021, February 8). Simple and Multiple Linear Regression [Video]. https://www.youtube.com/watch?
v=29rjWClT_3U
 SIMPLE LINEAR REGRESSION

o ľ»s goal of simpls li"saí ísgísssio" is ľo písdicľ ľ»s :al"s of

dsps"ds"ľ :aíiabls bassd o" i"dsps"ds"ľ :aíiabls.

o ľ»s gísaľsí ľ»s li"saí íslaľio"s»ip bsľwss" ľ»s i"dsps"ds"ľ

:aíiabls a"d ľ»s dsps"ds"ľ :aíiabls is, ľ»s moís acc"íaľs is
ľ»s písdicľio".

o W»aľ makss li"saí ísgísssio" a powsíf"l sľaľisľical ľool is

ľ»aľ iľ allows ľo q"a"ľif; b; w»aľ q"a"ľiľ; ľ»s
ísspo"ss/dsps"ds"ľ :aíiabls :aíiss w»s" ľ»s
sxpla"aľoí;/i"dsps"ds"ľ :aíiabls i"císasss b; o"s ""iľ.
15
Classified as Internal
 SIMPLE LINEAR REGRESSION
Example: A teacher wants to know if there is a relationship between the amount of time her students spent
working on a social studies report and the grade each student received. She surveyed 10 students and recorded the
data below.
1. CONDUCT A SCATTER PLOT 2.COMPUTE FOR X & Y VALUES
No. of
Stu- ANALYSIS
hours Gra No. of
dent Stu- Grade
worked hours XY X^2 Y^2
dent (Y)
worked (X)

1 5 1 5 90 450 25 8100
2 3 2 3 80 240 9 6400
3 3.5 3 3.5 80 280 12.25 6400
4 4 1 60 60 1 3600
5 5 4.5 90 405 20.25 8100
6 1 70 70 1 4900
6
7 3 75 225 9 5625
7
8 4 85 340 16 7225
8
9 2 70 140 4 4900
10 2.5 75 187.5 6.25 5625
Σ 29.50 775.00 2,397.50 103.75 60875
 SIMPLE LINEAR REGRESSION
Example: A teacher wants to know if there is a relationship between the amount of time her students spent
working on a social studies report and the grade each student received. She surveyed 10 students and recorded the
data below.
3. Find the Values of a & b.

(775)(103.75)-(29.50(2,397.50) 10(2,397.50)-(29.50)(77
a = 10(103.75)-(29.50)^2 b = 10(103.75)-(29.5

a9,680.00
= b =
1,112.50
167.25 1 Therefore, y = 57.88 + 6.65x
a = 57.88 b

What if the student made her report in SC for What if a student wanted to get a grade of 95? How
6 hours? What would be her estimated many hours should she spend making her report?
grade?
Answer: 5.58 hours
Answer: 97.78% 95 = 57.88 + 6.65x
Y = 57.88 + 6.65 (6) 95-57.88 = 6.65x
X = 5.58
 SIMPLE LINEAR REGRESSION
 Let's say a hospital ask you to give them an estimate based on Estimated length of age
the age of the person of how long the person will stay in the stay
hospital after a surgery.

ỹ=a+
bx
ỹ = 1.2 + 0.14x
ỹ = 1.2 + 0.14 (33)
ỹ = 5.82 days

Formula to calculate value of a & b is as follows:

Source: DATAtab. (2021, February 8). Simple and Multiple Linear Regression [Video].
https://www.youtube.com/watch?v=29rjWClT_3U
SIMPLE LINEAR
REGRESSION

Regression error

ỹ = a + bx + ε -is the
difference
between the
true value and
the estimated
value

Source: DATAtab. (2021, February 8). Simple and Multiple Linear Regression [Video].
https://www.youtube.com/watch?v=29rjWClT_3U
 MULľIPLE LINEAR REGRESSION

o U"liks simpls li"saí ísgísssio", m"lľipls li"saí ísgísssio"

ca" i"cl"ds ľwo oí moís i"dsps"ds"ľ :aíiablss.
o ľ»s goal is ľo ssľimaľs o"s :aíiabls bassd o" ss:síal oľ»sí
:aíiablss.
o ľ»s :aíiabls ľo bs ssľimaľsd is callsd ľ»s dsps"ds"ľ
:aíiabls (x-cíiľsíio"). ľ»s :aíiablss ľ»aľ aís "ssd foí
písdicľio" aís callsd i"dsps"ds"ľ :aíiablss (;-písdicľoís).
o M"lľipls li"saí ísgísssio" is ofľs" "ssd i" smpiíical social
ísssaíc». I" boľ» aísas, iľ is of i"ľsíssľ ľo fi"d o"ľ w»aľ
i"fl"s"cs diffsís"ľ facľoís »a:s o" a :aíiabls.

20
Classified as Internal
 MULľIPLE LINEAR REGRESSION

SIMPLE LINEAR REGRESSION MULTIPLE LINEAR REGRESSION

ỹ = a + bx

 The coefficients are interpreted similarly to the linear regression equation.

 If all independent variables are 0, the value a is obtained.

 If the independent variable changes by one unit, the associated coefficient b indicates by
how much the dependent variable changes.

Source: DATAtab. (2021, February 8). Simple and Multiple Linear Regression [Video].
https://www.youtube.com/watch?v=29rjWClT_3U
 USE IN ORGANIZAľION

o I" ľ»s fisld of b"si"sss, ísgísssio" is widsl; "ssd.

«"si"sssms" aís i"ľsíssľsd i" písdicľi"g f"ľ"ís
píod"cľio", co"s"mpľio", i":ssľms"ľ, píicss, píofiľs,
salss, sľc. So, ľ»s s"ccsss of a b"si"sssma" dsps"ds
o" ľ»s coííscľ"sss of ľ»s :aíio"s ssľimaľss ľ»aľ »s is
ísq"iísd ľo maks. Iľ is also "ssd i" sociological sľ"d;
a"d sco"omic pla""i"g ľo fi"d ľ»s píojscľio"s of
pop"laľio", biíľ» íaľss, dsaľ» íaľss, sľc.

22
Classified as Internal
 HYPOľHESIS i⭲ ísgísssio⭲

o Píoblsm Sľaľsms"ľ: Is ľ»sís a íslaľio"s»ip bsľwss"

sd"caľio" a"d i"coms i" ľ»s ""mbsí of c»ildís" i" a
famil;?

o N"ll H;poľ»ssis: ľ»sís is "o íslaľio"s»ip bsľwss"

sd"caľio" a"d i"coms of ísspo"ds"ľs a"d ľ»s ""mbsí
of c»ildís" i" familiss.

2«
Classified as Internal
LIľERAľURE SAMPLES-REGRESSION
 26
Classified as Internal
 Linear Regression
Example Problem
Simple Linear regression equation
Y = Β0 + Β1X

Classified as Internal
Example 1:
You have to study the relationship between monthly e-commerce
sales and online advertising costs. You have the survey results for 7
online stores for the last year.

Online Store Monthly E-commerce Sales (in 1000s) Y Online Advertising Dollars (1000 s) X

1 368 1.7

2 340 1.5

3 665 2.8

4 954 5

5 331 1.3

6 556 2.2

7 376 1.3

Classified as Internal
 1. Method #1- Scatter Chart with a Trendline

a. Sslscľ ľ»s ľwo col"m"s of ľ»s daľassľ (x a"d ;), i"cl"di"g »sadsís b. Click o" ‘I"ssíľ’ a"d sxpa"d ľ»s díopdow" foí ‘Scaľľsí
C»aíľ’ a"d sslscľ ‘Scaľľsí’ ľ»"mb"ail (fiísľ o"s)

Classified as Internal
 c. Now a scatter plot will appear, and under the Design, look for Select
Data ,and select data source will appear, Click the Add and then input a d. Now in the ‘Format Trendline’ pane on the right, select
Series name and the X-value and Y-value(don’t include the label). Then ‘Linear Trendline’ and ‘Display Equation on Chart’
Click OK. And another OK. To do this, right-click on any data point
and select ‘Add Trendline.

Classified as Internal
 E. Select ‘Display Equation on Chart and Display R-
squared value on chart’.

Classified as Internal
 Msthod 2: Usi⭲g Data A⭲al;sis

A ísgísssio" dialog box will appsaí. Sslscľ ľ»s I"p"ľ Y ía"gs a"d I"p"ľ X
ía"gs. I" ľ»s cass of m"lľipls li"saí ísgísssio", ws ca" sslscľ moís
col"m"s of i"dsps"ds"ľ :aíiablss.
C»sck ľ»s ‘Ḻabsls’ box ľo i"cl"ds »sadsís.
C»ooss ľ»s dssiísd ‘o"ľp"ľ’ opľio".
a. Click on ‘Data Analysis’ in the ‘Data’ tab and select Regression
Sslscľ ľ»s ‘íssid"als’ c»sckbox a"d click ‘OK. (Opľio"al)

Classified as Internal
 Now our regression analysis output will be created in a new
worksheet, stating the Regression Statistics, ANOVA, residuals
and coefficients.

Classified as Internal
 Steps to perform this linear regression in SPSS.

Step 2: Perform linear regression.

Step 1: Enter the data. Click Analyze > Regression > Linear... on the top menu, as shown
below:

Classified as Internal
 Drag the variable Monthly E-commerce into the box labelled Dependent.
Drag the variables Online Advertising into the box labelled Independent(s).
Then click OK.

Classified as Internal
 Rssult

Classified as Internal
 Example 2:
You have to examine the relationship between the age and price for used cars
sold in the last year by a car dealership company.
Car Age (in years) Price (in dollars)
4 6300
4 5800
5 5700
5 4500
7 4500
7 4200
8 4100
9 3100
10 2100
11 2500
12 2200

Classified as Internal
 Steps to perform this linear regression in SPSS.

Step 2: Perform linear regression.

Step 1: Enter the data. Click Analyze > Regression > Linear... on the top menu, as shown below:

Classified as Internal
 Drag the variable Car Price into the box labelled
Dependent. Drag the variables Car age the box labelled
Independent(s). Then click OK.

Classified as Internal
 Rssult:

Classified as Internal
 Rssult usi⭲g Excsl

Classified as Internal
 MULTIPLE
REGRESSION

formula y = b0 + b1*x1 +
b2*x2 + …..

Classified as Internal
Exampls Píoblsm:

The ABC Corporation is opening new retail sales outlets and they want to staff these stores with
employees most likely to be successful at selling the products. To meet this goal, ABC
decides to study the sales staff at existing stores to determine if intelligence and extroversion (i.e.,
a friendly and outgoing personality) predict the sales performance of current employees. ABC's
logic is that if intelligence and extroversion predict sales performance, then a good strategy for
new stores is to hire intelligent extroverts for the sales positions.

To conduct the study, all current retail sales employees at existing stores take psychological tests
designed to measure intelligence and extroversion. Also, past sales performance data is checked
for each employee. In the end, there are three scores for each salesperson:
1. an intelligence score (on a scale of 50-low intelligence to 150-high intelligence),
2. an extroversion score (on a scale of 15-low extroversion to 30-high extroversion), and
3. sales performance is expressed as the average dollar amount sold per week.

Classified as Internal
 Sales Person Intelligence Extroversion $ Sales/Week
1 89 21 2625
2 93 24 2700
3 91 21 3100
4 122 23 3150
5 115 27 3175
6 100 18 3100
7 98 19 2700
8 105 16 2475
9 112 23 3625
10 109 28 3525
11 130 20 3225
12 104 25 3450
13 104 20 2425
14 111 26 3025
15 97 28 3625
16 115 29 2750
17 113 25 3150
18 88 23 2600
19 108 19 2525
20 101 16 2650

Classified as Internal
 Msthod 2: Usi⭲g Data A⭲al;sis

A ísgísssio" dialog box will appsaí. Sslscľ ľ»s I"p"ľ Y ía"gs (Salss/Wssk) a"d I"p"ľ X ía"gs
(i"ľslligs"cs a"d Exľío:sísio"). I" ľ»s cass of m"lľipls li"saí ísgísssio", ws ca" sslscľ
a. Click o" ‘Daľa A"al;sis’ i" ľ»s ‘Daľa’ ľab a"d sslscľ ksgísssio" moís col"m"s of i"dsps"ds"ľ :aíiablss C»sck ľ»s ‘Ḻabsls’ box ľo i"cl"ds »sadsís.
C»ooss ľ»s dssiísd ‘o"ľp"ľ’ opľio".

Classified as Internal
 DAľA ANALYSIS RESULľ IN
EXCEL

Classified as Internal
 Steps to perform this multiple linear regression in
SPSS.

Sľsp 1: E"ľsí ľ»s daľa. Sľsp 2: Psífoím m"lľipls li"saí ísgísssio".

Classified as Internal
 Díag tks :aíiabls Salss/!ssfi i⭲to tks box labsllsd
Dsps⭲ds⭲t. Díag tks :aíiablss i⭲tslligs⭲cs a⭲d
Extío:sísio⭲ i⭲to tks box labsllsd I⭲dsps⭲ds⭲t(s). ľks⭲
clicfi OK.

Classified as Internal
 Stsp «: I⭲tsípíst tks output.

Classified as Internal
 ľHANK YOU!!!!

Classified as Internal