SPSS STEPS

Thanushki Tharika

1|Page
 ANALYSIS OF DEMOGRAPHIC FACTORS
✓ Go to Variable view in SPSS

✓ Check the Part A in the Questionnaire (General Questions) and enter names to
the SPSS

Name – Question Name (*When enter names not use spaces between words. Use Underscore)
First enter 6 names and then fill other columns
Type – Selecting Data Type
• If the question is categorical (E.g.: Age – 20-25, 25-30), data type is String.
• If the question is a specific number then the data type is Numeric.
Label – Copy paste the Full name of the Question (Entire Question) in the questionnaire
Values – Give the value labels to each question
Eg: Value – 1 Label – below 25 Add Then enter all 5 Categories and OK

2|Page
Missing – After collecting data there shouldn’t be any missing value.
If there is any missing value there are 2 options to do.
Option 1 – Remove that respondent
Option 2 – Give an answer by the researcher, but it is not ethical.
So, give a random number like 99999999. To give that click on … in the cell and go to Discrete
Missing values and then enter 9999999 in the box and Ok.

✓ Then enter the data

Go to Data View

3|Page
 ✓ Then go to the data file (Excel Sheet) Demographic Data

Then Ctrl + F Then fill the table

E.g.: When take the “Your Age” in the Questionnaire
Find - Below 25 years Replace – 1 Replace All OK
Find – 26-30 Years Replace – 2 Replace All OK

4|Page
✓ Then find all the data of 6 demographic questions of the questionnaire like above.
✓ Then it looks like this.

✓ Then copy all that data and go to SPSS Then Right Click on the first cell and
Click on paste
✓ To view data correctly Click on the Red circled icon below, after that all data will look
like below.

5|Page
Descriptive analysis / Demographic Analysis
E.g.: Analysis of Age
✓ Analyze Descriptive Statistics Frequencies

6|Page
 ✓ Then it will appear like following.

You have the list here which show all the labels. If you don’t want the label and you want the
name, then right click on the variable and select the “Display Variable Names”.

✓ Then select the variable you want and click on the arrow in the middle (Yellow Circle)
and Then Click the “Statistics” (Red Circle)

7|Page
✓ Then you can select any statistic like mean, mode, median etc. Here as an example we
select the “Mean”. After selecting “Mean” then click on “Continue”.

✓ Then Click on “Charts”.

✓ Then you can select any chart like

Bar Charts, Pie Charts or Histograms.
Here we select Pie Charts. And the if
you want percentages, then select
“Percentages”. Then Click
“Continue”

8|Page
✓ Then put a Tick on “Display Frequency Tables” and Click “OK”.

✓ Then you will open a separate window (Out Put Window) with Frequency Table and a
pie Chart like below. You also can save this separately.

✓ Then you copy that Pie Chart and Table to your word file (Chapter 4). (See the Word
File)
Figure 4.1: Pie chart
for age distribution

9|Page
 ✓ If you want to edit the pie chart or put percentages on the Pie Chart then Double Click
on it. Then you will get the “Chart Editor” like below.

✓ If you want to change colors or patterns in the Pie Chart, then put a single click on the Pie
Chart and then that area will be selected. Then you will see the “Property Window” next to
it.

10 | P a g e
✓ You can use any pattern from here.

✓ You can change colors by using click on “Fill” Select the color you want
Apply

11 | P a g e
✓ If you want to add percentages on the Pie Chart. Click on the Icon “Data Label Mode”.

✓ When you click on the “Data Label Mode”, the curser is now representing like an ICON.
✓ Then just come on to your Pie Chart and give a single click. Then you will see the
percentage.

To save this file. Go to,

12 | P a g e
File Save Give the location where you want to save the file

Then give a name to the file Save

✓ Then you can follow the above steps and do data analyze for the other variables also.
To do that click on “Restart”.

13 | P a g e
E.g.: Analysis of Gender

Sex

Frequency Percent Valid Percent Cumulative

Percent

Male 50 57.5 57.5 57.5

Valid Female 37 42.5 42.5 100.0

Total 87 100.0 100.0

E.g.: Analyze of Service

✓ Analyze Descriptive Statistics Frequencies Service Charts

14 | P a g e
✓ This time select Bar Charts and Click “Continue”. Then Click on “OK”.

15 | P a g e
Cross Tabs
✓ Cross tabulation is a representation of more than two demographic questions in one
table/chart.
✓ Analyze Descriptive Statistics Crosstabs

Then select the two variables you want. Here we selected Age and Gender. Then go to
“Cells”.

16 | P a g e
✓ Then cell display will open like below. Select the “Row”, “Column” and “Total”.
✓ Then Click “Continue” Ok

✓ Then it will open a new window like this.

✓ These types of big tables put to Annex, don’t put to analysis chapter.
✓ Draw another table in the Analyzing chapter to show above details. (Refer the Word
file)

17 | P a g e
 ✓ If you want a graph for this. Following is the path.
Graphs Legacy Dialogs Bar (You can select any chart what you want)

✓ Select Clustered Define

18 | P a g e
✓ Select “% of cases” and then add two variables you want. Then Click on “OK”.

19 | P a g e
✓ Then a new window will open as below.

✓ This is called “Clustered Bar Diagrams”. (More than one variable)

✓ Earlier we have taken Simple Bar Diagrams where only have one variable.

✓ So those all are Descriptive Statistics.

20 | P a g e
 4.2 REGRESSION ANALYSIS
This is the most important part in the research. Conclusion or findings is entirely based on this
regression analysis.
✓ First go to the Likert Scale Questions in the Questionnaire.
All those 25 questions are related to the one of the factors in the conceptual frame work.
(Independent and Dependent Variables).
Here it has not given which question related to which variable so first find that. It has indicated
below.
EE = Employee Engagement
P = Promotion
S = Salary/ Incentives
Independent Variables
R =Recognition
TD = Training & Development
TI = Turnover Intention Dependent Variable

No. Statements SA A N DA SDA

Manager/AUH value my opinions and suggestion when allocate employee
EE1 5 4 3 2 1
for the sections.
P2 Career Development programs available in the organization.

S3 Company provide reasonable amount of performance incentives.

TI4 If I had a different job my benefits would probably improve.

R5 Manager allocates newcomers to train under my supervision.

Benefits offered by my organization are a key reason why people join the
TI6
organization.
Training and Development programs give me motivation for career
TD7
advancement.
TI8 Currently I am actively searching for other employment opportunities

EE9 Manager / AUH ask me to handle complain of customer in my section.

P10 Managers help me to groom my knowledge and skill for next promotion

TD11 Training and development programs based on my performance evaluation.

S12 Organization is fair for overall benefits of employees.

R13 There is a clear reward system based on performance in organization

S14 Based allowances, Staff purchasing facility make my life smoother.

21 | P a g e
S15 I am received adequate salary for my effort.

R16 I am recognized by superior for performance of my subordinates.

TD17 Training programs help me to perform well.

TI18 I would move to another organization within the next 12 months

R19 I am recognized by Manager/ AUH/ OM when achieve my budget

P20 Job training and Career Advancement are given in fair manner.

EE21 Manager/AUH allow to me to make morning briefing.

TI22 My expectations were met after i joined the organization.

P23 There is promotion and appraisal procedure in the organization

S24 Medical benefit makes my family and my life healthier

TD25 Training programs are relevant for my task.

*All the independent and dependent variables should have questions in the
Questionnaire.
*All the independent variables should have at least 3 Questions and maximum 5
Questions in the Questionnaire.
✓ Then go to the “Variable View” in SPSS and enter all above Variables as follows.
You should enter all the variables according to the order in the conceptual frame work.

22 | P a g e
Type – Numeric (For all Likert Scale Questions, data type should be “Numeric”)
Decimal – Reduce decimals to “0”, if you don’t like decimals.
Label – Copy Paste the Entire Question in the Questionnaire.
Values –
E.g.: Value – 1 Label – “Strongly disagree” Add
Value – 2 Label – “Disagree” Add
✓ Likewise enter all the 5 Likert Scales and finally Click on “OK”.
You don’t want to type those 5 Likert scale values in every question again and again, you can
copy paste this to other questions.

✓ Missing – Enter 999999

23 | P a g e
✓ Measure – Select the “Scale”.

✓ Then go to the Excel Data Sheet Likert Scale Data

24 | P a g e
 ✓ Then select the data under the correct question number and copy it.
E.g. Employee engagement represented by Question Numbers – EE_1, EE_9, EE_21

✓ Then go to the SPSS Data View

✓ Right Click on the First cell under EE_1(Yellow Colored Circle) and paste it.
✓ Enter all the data like that and finally it will look like below.

25 | P a g e
RELIABILITY ANALYSIS
If a scale producing consistent results, we can identify such measurement as a reliable
measurement. The reliability analysis explains the extent to which a scale is producing
consistent results.
✓ Analyze Scale Reliability Analysis

✓ Right Click on the Question Display Variable Name

26 | P a g e
 ✓ Then send the Questions of the first independent variable to the Items Cage by clicking
the Arrow in the middle.

Then go to statistics Put ticks to the “Item”, “Scale” & “Scale if item deleted”
Continue Ok

27 | P a g e
✓ Then a new window will open like below.

✓ There are 5 tables as below.

28 | P a g e
 ✓ Ignore the first table and Copy the second table to your Analysis.
Reliability Statistics for Employee Engagement

Reliability Statistics

Cronbach's N of Items
Alpha

.862 3

That is called Cronbach’s Alpha. That is the measurement of the “Reliability”.

Rule for Reliability

Cronbach’s Alpha should be above O.7

• If Cronbach’s Alpha is under 0.7, then our questions are wrong or our answers are
wrong or something has gone wrong somewhere.

• So, we need to correct it. Without correcting it we cannot run the Regression.

✓ Above Cronbach’s Alpha is for Employee engagement and it is 0.862. So, it is above
0.7, that means all that 3 questions are correct. So, it all giving consistent results.

✓ Then measure the “Reliability” for other variables as above.

29 | P a g e
 ✓ We take Recognition as the second variable.
Reliability Statistics for Recognition

✓ Then follow above steps and obtain the Cronbach’s Alpha.

Reliability Statistics

Cronbach's N of Items
Alpha

.587 4

• So here there is a problem because Cronbach’s Alpha is 0.587. It is less than the 0.7.
• So, questions for “Recognition” has an issue.

Reliability Statistics for Salary Reliability Statistics for Reliability Analysis for
Training and Development Promotion
Reliability Statistics
Reliability Statistics Reliability Statistics
Cronbach's N of Items
Cronbach's N of Items Cronbach's N of Items
Alpha
Alpha Alpha
.871 5
.884 4 .820 4

30 | P a g e
Reliability Statistics for Turnover Intention (Dependent Variable)

Reliability Statistics
• So here there is a problem in the
Cronbach's N of Items
dependent variable. It comes as a negative
Alpha
value due to a negative average
-.654 5 covariance among the items. This violates
a. The value is negative due to the reliability model assumptions. That
a negative average covariance means entire questions for turnover
among items. This violates intention are wrong.
reliability model assumptions.
You may want to check item
coding.

✓ Take all of these above tables to Annex (Refer the Annex Doc). Do not put in the
Chapter- Analysis.
✓ In the chapter you create a table, summarizing above all to depict clearly as below.

The variable Cronbach’s Reliability Conclusion Item-if-

Alpha deleted
Employee 0.862 0.862> 0.7 Reliable
Engagement
Recognition 0.587 0.587<0.7 Not reliable If item is
deleted still
value is
0.639 and
not reliable
salary 0.871 0.871>0.7 Reliable
Training and 0.884 0.884>0.7 Reliable
development
Promotion 0.820 0.820>0.7 Reliable
Turnover 0.790 0.790>0.7 Reliable There will
intention three
question to
be removed

• Add extra column to the table because there are two variables that is not reliable and to
make them reliable, we have to remove some questions and adjust it. That adjustments
should clearly show in the table.
• To make the variables “Recognition” reliable run the reliability analysis again.

✓ Then copy the bigger table (Item – Total Statistics) to the Annex.

31 | P a g e
 Item-Total Statistics

Scale Mean if Scale Variance Corrected Item- Cronbach's

Item Deleted if Item Deleted Total Alpha if Item
Correlation Deleted

Manager allocates 9.40 5.011 .350 .548

newcomers to train under
my supervision.
There is a clear reward 10.15 6.477 .207 .639
system based on
performance in organization
I am recognized by superior 9.47 5.345 .588 .358
for performance of my
subordinates
I am recognized by 9.18 6.291 .407 .498
Manager/ AUH/ OM when
achieve my budget

Refer the last column of the table. If we delete any item, look anything which is higher
than 0.7 is available in this column. But here there is no any factor above 0.7. The highest
one is 0.639. So, we have to delete that entire question from the SPSS. It is deleting because
the problem is with the question that has the highest number. That means if we delete
that entire question, we can higher the statistic to the 0.639. Before deleting from SPSS,
you run the reliability analysis again without that question and check whether it can
change above 0.7.

32 | P a g e
✓ Then remove that question (R2) and run the reliability analysis for the recognition
again.

Reliability Statistics

Cronbach's N of Items
Alpha
.639 3

33 | P a g e
 • Then you can see value has been increased to 0.639. But still it is not reliable. So,
there is no any advantage of deleting that question.
✓ Then write this in the last column of the table which is in the chapter 4 as “If item is
deleted still value is 0.639 and not reliable”.

The variable Cronbach’s Reliability Conclusion Item-if-

Alpha deleted
Employee 0.862 0.862> 0.7 Reliable
Engagement
Recognition 0.587 0.587<0.7 Not reliable If item is
deleted still
value is
0.639 and
not reliable
salary 0.871 0.871>0.7 Reliable
Training and 0.884 0.884>0.7 Reliable
development
Promotion 0.820 0.820>0.7 Reliable
Turnover 0.790 0.790>0.7 Reliable There will
Intention three
question to
be removed

• If assume we get above the 0.7, we can delete the entire question from the SPSS and
run the reliability analysis again and report in this above table as “If item deleted then
it can be increased”.
✓ Then we take “Turnover Intention”, all the questions about Turnover Intention are
incorrect because it has a negative value.
✓ Then refer the questions regarding Turnover Intention in the questionnaire again and
remove the questions what you feel as wrong.
✓ Then run the reliability again by removing those questions.

34 | P a g e
 Reliability Statistics

Cronbach's N of Items • Still there is a problem because it has a negative value

Alphaa due to negative average co variance.
-.810 3

a. The value is negative due to

a negative average covariance
among items. This violates
reliability model assumptions.
You may want to check item
codings.

✓ Then run the reliability analysis by deleting questions.

• By deleting TI_2, TI_4 and TI_5 we can obtain a Cronbach’s Alpha above 0.7.

Reliability Statistics

Cronbach's N of Items
Alpha

.790 2

• So, TI_1 (If I had a different job my benefits would probably improve) and the TI_3
(Currently I am actively searching for other employment opportunities) are the two
acceptable questions regarding the Turnover Intention.
✓ Then report that in the table in the Chapter 4.
• Dependent Variable should be reliable definitely.

35 | P a g e
 ✓ Then delete the 3 questions that is not acceptable (TI_2, TI_4 & TI_5) from the SPSS.
✓ To delete Right click on the number and click on “Clear”.

Then after deleting questions it will looks like that.

36 | P a g e
 CORRELATION ANALYSIS
Correlation mean what is the relationship between independent variables and the dependent
variable.
The correlation analysis describes the relationship between two continuous variables in here,
the relationship between Employee engagement (Independent) and Turnover intention
(Dependent) will be analyzed as follows.
Go to the SPSS,
Transform Compute Variables

✓ Then you can have a table like follows. Fill the Target Variable –
Employee_Engagement (Don’t keep spaces use underscore)

✓ Then go to here and right

click and click on “Display
Variable Name”.

37 | P a g e
✓ Then fill the Numeric Expression Cage
✓ Type Mean then open a bracket and Double Click on “EE_1” or you can also click on
“EE_1” and then click on the arrow head.
✓ Then put a coma and double click on “EE_2” then again put a coma and double click
on “EE_3” and close the brackets and Click on “OK”.
✓ Then you didn’t get any table. You will get a Execute Statement as follows.

COMPUTE
Employee_Engagement=Mean(EE_1,EE_
2,EE_3.
EXECUTE.

38 | P a g e
 ✓ Then you continue the process.

✓ Then do the process for the “Turnover Intention” which is the dependent Variable.

✓ Transform Compute Variables

✓ You can Reset the table and fill the cages as below.
✓ Remember don’t use spaces, use underscore.

Then also you can see an Execute Statement as follows.

COMPUTE Employee_Engagement=Mean(EE_1,EE_2,EE_3).
EXECUTE.
COMPUTE Turnover_Intention=Mean(TI_1,TI_3).
EXECUTE.

39 | P a g e
✓ Then go to the SPSS and Go to the “Data View”. Then you can see data has been filled.

✓ Then we are going to run the “Run the Correlation”.

✓ Analyze Correlate Bivariate

40 | P a g e
✓ Then you can have “Bivariate Correlation” Table as follows.

✓ Right Click on here and Click on “Display Variables Names”.

✓ Then go down of this cage and Click on “Employee_Engagement” and then Click on
“Turnover_Intention”.

41 | P a g e
✓ Then it will look like follows. Put a “Tick” in “Pearson” and Then Click on “OK”.

✓ Then you will open a new window like this.

42 | P a g e
 ✓ Then copy that table to “Annex”. (Refer the Annex Doc)
Correlations

Employee_Eng Turnover_Intenti
agement on

Pearson Correlation 1 -.280**

Employee_Engagement Sig. (2-tailed) .009

N 87 87
Pearson Correlation -.280** 1

Turnover_Intention Sig. (2-tailed) .009

N 87 87

**. Correlation is significant at the 0.01 level (2-tailed).

✓ Then Draw a table in your Chapter 4 to summarize the data as follows.

Table 1.2 Analysis of correlation between employee engagement and turnover intention

Dependent Independent Variable Pearson Significance of

Variable Correlation the Relationship
Turnover Invention Employee Engagement -0.280 0.009

According to Table 1.2 the Pearson Value is Negative. It says that there is a negative
relationship between employee engagement and turnover intention. Also, the relationship is
28% that means when employee engagement increased turn over intention can be decreased.
Since the significance value is less than 0.05, this relationship is significant.
Significance Value Cannot be a negative number. It should be less than 0.05.

✓ Then run the correlation for the next variable.

✓ Next going to run the correlation for the “Recognition”.

✓ Go to, Transform Compute Variables

43 | P a g e
 • We don’t want to make “Turnover Invention” which is the dependent variable because
it has already done.

✓ Then Go to,
Analyze Correlate Bivariate
✓ Then select “Recognition” and “Turnover Intention” and Tick the “Pearson” and then
Click on “OK”.

44 | P a g e
 ✓ Then you will get a table like this and copy this into the Annex.

Correlations

Recognition Turnover_Intenti
on

Pearson Correlation 1 -.518**

Recognition Sig. (2-tailed) .000

N 87 87
Pearson Correlation -.518** 1

Turnover_Intention Sig. (2-tailed) .000

N 87 87

**. Correlation is significant at the 0.01 level (2-tailed).

✓ Then write about this in the Chapter 4.

Table 1.4 Analysis of correlation between Salary and turnover intention

Dependent Variable Independent Variable Pearson Correlation Significance of the

Relationship
Turnover Invention Salary -0.733 0.000

There is a negative relationship between Salary and turnover intention. Also, the relationship
is 73% that means when Salary increased turn over intention can be decreased.

✓ Likewise run the correlation for other variables too.

45 | P a g e
 REGRESSION ANALYSIS

• Regression is the final part of the analysis and it is the most critical part in a research.
• If the Regression is not significant, non of your conclusions are accepted.
• Regression says whether the conceptual framework is correct or incorrect.
• The regression analysis explains the predictive power of the model defined. In this we
can test each of the hypothesis defined earlier. If there are more than 1 independent
variables, we need to run a multiple regression (e.g.; The model here). If you have only
one independent variable then the analysis would be a regression. In the regression
analysis we can see the overall predictive power and the predictive power of each
independent variable on the dependent variable.

• This is somewhat complicated because we have overall predictive power and individual
predictive powers.

✓ Go to SPSS to run the Regression.

Analyze Regression Linear

46 | P a g e
 ✓ Right Click on the
questions and select the
“Display Variable Names”.

✓ To enter the “Dependent” variable go down in the variables cage and select the
“Turnover_Intention”. Don’t select TI_1, select the variables you newly created as
circled below. Then click on the “Arrow”.

47 | P a g e
✓ Then again go down in the variables cage and select all the independent variables one
by one as per the order in the conceptual framework and send to the “Independent(S)”
cage by click on the “Arrow”.

✓ Then go to “Statistics”.

✓ Then it will open a “Linear Regression: Statistics” table and already it has been selected
“Estimates” and “Model Fit”. Then you put a “Tick” on “Confidence Intervals”. Then
click on “Continue” and then “OK”.

48 | P a g e
 ✓ Then you will open a new window like follow. There are 4 tables.

Variables Entered/Removeda

Model Variables Variables Method

Entered Removed
✓ Copy the first table and paste to
Training_and_D . Enter
the “ANNEX”. Check whether all the
evelopment,
Employee_Eng
variables are there.
1 agement,
Salary,
Recognition,
Promotionb

a. Dependent Variable: Turnover_Intention

b. All requested variables entered.

Model Summary

Model R R Square Adjusted R Std. Error of the

✓ Paste the second table also
to the ANNEX.
Square Estimate

1 .814a .662 .641 .76184 There 2 required information in

a. Predictors: (Constant), Training_and_Development, this table (Model Summary).
Employee_Engagement, Salary, Recognition, Promotion
One is “R Square” and other
one is “Adjusted R Square”.

49 | P a g e
 • If R Square value is greater than 0.5, it explains a good model fit.
• If you obtain a high R Square value like 0.8 or 0.9 then it is a perfect fit.
• If you obtain a 0% then it explains none of the variability, that means your model
cannot use.
• If you get 100% (1.0) then the model is perfectly fitted to the Regression Analysis.
That means you have explained all of the variance that there is to explain. And
also, you can get R Square value = 1 if you have a number of predicting variables
equal to the number of observations.
• The adjusted R-squared is a modified version of R-squared that has been adjusted
for the number of predictors in the model. The adjusted R-squared increases only
if the new term improves the model more than would be expected by chance. It
decreases when a predictor improves the model by less than expected by chance.
The adjusted R-squared can be negative, but it’s usually not. It is always lower
than the R-squared.
• If “Adjusted R Square” is less than the “R Square”, that means if we add one more
independent variable the productive power become low. So, no need to put another
variable because R Square number is normal, already model is functioning well.

ANOVAa

Model Sum of Squares df Mean Square F Sig.

Regression 92.143 5 18.429 31.751 .000b

1 Residual 47.012 81 .580

Total 139.155 86

a. Dependent Variable: Turnover_Intention

b. Predictors: (Constant), Training_and_Development, Employee_Engagement, Salary,
Recognition, Promotion

✓ Copy the “ANOVA” Table also to the ANNEX.

• In this table there are lot of information but we only need the “Significance” Value.

• That is called “Model Significance”.

• That is the most important value to determine whether this research is important
or not. That means from that value we can know the entire model of the research
is correct or incorrect.

• Model Significance value should be less than 0.05.

50 | P a g e
 ✓ Then go the Chapter 4 and report the summary of the Regression as follows.
Summary of Regression

Model Summary

R-square 0.662

Adjusted R-square 0.641

Model 0.000
Significance

“R Square is the percentage of the response of the response variable variation that is explained
by a linear model” (Reuf,2017). When R-square is high the model fit will increase. In here the
R-square is 0.662 which is greater than 0.5 explain a higher model fit. The R-square value also
explain that all the independent variables together explain the dependent variable by 66%. That
means all the independent variables together impact on / contribute to turnover intention 66%.
The adjusted R-square explains that “Whether additional input variables are contributing to the
model” (…). Since adjusted R-square is 0.641 it says that when a new variable is added
productive power will decrease.
In the ANOMA Table the model significance is 0.000 which is less than 0.05, the model is a
significant model.

Hypothesis Testing
In the Hypothesis Testing, to Accept or Reject we need to interpret both the “Beta
Coefficient” and “Significance”.
First look at the Significance and If the significance value is less than 0.05, we can
interpret and Accept the hypothesis if the sign (What we put in hypothesis earlier –
Positive or Negative) is correct in the Beta Coefficient.

✓ Then copy the “Coefficient” Table to the ANNEX.

✓ When copy it, it cut the page margins.

Then select the table Layout Autofit Autofit to Window

51 | P a g e
 Coefficientsa

Model Unstandardized Standardized t Sig. 95.0% Confidence

Coefficients Coefficients Interval for B

B Std. Beta Lower Upper

Error Bound Bound

(Constant) 6.827 .410 16.642 .000 6.011 7.643

Employee_Engagement .024 .106 .020 .223 .824 -.188 .235

Recognition -.035 .173 -.021 -.203 .840 -.379 .309

1
Salary -.497 .109 -.438 -4.565 .000 -.714 -.280

Promotion -.157 .174 -.102 -.901 .370 -.502 .189

Training_and_Development -.500 .162 -.372 -3.091 .003 -.822 -.178

a. Dependent Variable: Turnover_Intention

52 | P a g e
 ✓ Then go to the Chapter 4
Hypothesis Testing
Hypothesis Testing Beta Significance Accept/Reject
Coefficient
HI: The turnover intention is 0.024 0.824 Reject
negatively related to the employee
engagement among middle level
employees in ABC hotel corporation
H2: The turnover intention is -0.035 0.840 Reject
negatively related to the Recognition
among middle level employees in ABC
hotel corporation
H3: The turnover intention is -0.497 0.000 Accept
negatively related to the Salary among
middle level employees in ABC hotel
corporation
H4: The turnover intention is -0.157 0.370 Reject
negatively related to the Promotion
among middle level employees in ABC
hotel corporation
H5: The turnover intention is -0.500 0.003 Accept
negatively related to the Training and
Development among middle level
employees in ABC hotel corporation

HI: The turnover intention is negatively related to the employee engagement among middle
level employees in ABC hotel corporation
According to significance in regression results for employee engagement, the value is greater
than 0.05(0.824>0.05). Therefore, H1 should be rejected. Hence there is no impact from
employee engagement on turnover intention among middle level employees in ABC Hotel
Corporation. (We can accept the H0).
H2: The turnover intention is negatively related to the Recognition among middle level
employees in ABC hotel corporation
According to significance in regression results for Recognition, the value is greater than
0.05(0.840>0.05). Therefore, H2 should be rejected. Hence there is no impact from
Recognition on turnover intention among middle level employees in ABC Hotel Corporation.

53 | P a g e
H3: The turnover intention is negatively related to the Salary among middle level employees
in ABC hotel corporation
According to the significance results for salary, the value is less than 0.05(0.000<0.05). Also,
the Beta Coefficient is -0.479 which explains that there is 49.7% impact from salary/incentives
to middle level employee turnover intention in ABC Hotel Corporation.
Finding – Out of five variables, Salary is the most important factor which can really make an
impact on Employee Turnover Intention. This is a negative relationship as described therefore
if salary increases by 1% in this organization middle-employee turnover intention can be
decreased by 49%.
Findings comes from Accepted ones.
Likewise report about other two variables also.
Finding - 50% of Turnover Intention can be reduced if 1% of Training and development is
given.
So, in this research 2 main conclusions can be made.
The most contributing factor to the turnover intention is “Training and Development” which is
50%. And the second most contributing factor to the turnover intention is “Salary” which is
49%.
Other 3 factors are not contributing to the Turnover Intention.
However, this model is a significance model to predict “Turnover Intention”. As a model this
is accepted.

54 | P a g e