Anda di halaman 1dari 73

Transform Data into Knowledge 41 SPSS

Context: The objective of the assignment is to study the regression analysis of the validity co-efficient which we get in the correlation analysis of our sample. For the analysis and learning purpose we had taken data from the BBA-16 (A & B) of Graduate School of business at International Islamic University, Islamabad. We recorded there CGPA, Intermediate percentage, medium of instruction in Matric , intermediate institution, accounting1, accounting 2, cost accounting, English1, English 2 and oral communication. In the previous assignment we studied we the correlation and in this we will study the regression of these variables. We have made a comparison, of their CGPA, Inter-Percentage, MOIM, Public or Private Institution, Accounting Grades and Functional English Grades in the Variable View by giving them appropriate values. We study the following variables in our analysis:1) 2) 3) 4) 5) 6) CGPA of the students, ( Dependent Variable) Intermediate Percentage, ( Independent Variable) Medium of Instruction up to Matric, ( Independent Variable) Grades of Accounting 1, 2 and 3, ( Independent Variable) Grades of Functional English 1 and 2, ( Independent Variable) Public or Private Sector, (Independent Variable).

Transform Data into Knowledge 42 SPSS

Regression Models:-

As we have 6 variables but 1 is dependent variable. So we has five validity coefficient; which are as given below

The above given validity co-efficient we get from Correlation Matrix.

Transform Data into Knowledge 43 SPSS

Question: Which is the Predictor we should study first? The answer is that the validity co-efficient with highest value of significance will study first. The reason is because this has highest impact on the dependent variable. So we should study these independent variable on there importance. So we study Quantitative first as it has .526** level of significance. After it we study Verbal which has .488** level of significance. In the same we study MOIM, Inter-Percentage and Public or Private as they have .261, .081 and .020 level of significance respectively.

So we have to study five Regression models. Model # 1: =a+b1x1 (Quantitative) Model # 2: =a+b2x2 (Verbal) Model # 3: =a+b3x3 (MOIM) Model # 4: =a+b4x4 (Inter-Percentage) Model # 5: =a+b5x5 (Public or Private)

Transform Data into Knowledge 44 SPSS

QUNT Verbal MOIM Inter-pet Istit


Dependent Variable Independent Variables DVs

CGPA

IVs

Transform Data into Knowledge 45 SPSS

This is the simple model which we will study in this assignment. At first; When we open the data source file than this view will open, it has two views the Data view and variable view these are explained in the previous assignment

Transform Data into Knowledge 46 SPSS

Model 1 (CGPA and QNT) Scatter Dot We want to calculate the regression, for this purpose we shall go in Graphs then Legacy Dialogues and click on Scatter Dot. As shown in following dialogue box

Theses are the steps for finding out the Scatter Dot Matrix; 1. First go to Graphs 2. Click on Legacy Dialogue 3. Now click on Scatter Dot 4. Select Simple Scatter 5. Click Define 6. Select CGPA on Y-axis. 7. Select Quant on x-axis. 8. Click Ok 9. Simple Scatter is formed 10.Double click on it. 11.Click on Add to fit lines at Total. 12.Click on Linear and now click on apply. 13.A Linear curve is formed 14.Click on Add to Fit Line. 15.Click on Loess and then click on Apply. 16.Loess Curve is formed.

These are the general steps that would be used to find out the regression of all the variables. In order to draw a Simple Scatter we have to follow following steps shown in the fig. These entire steps are those which are mention earlier

Transform Data into Knowledge 47 SPSS

When we click on scatter Dot, the following Dialogue box will appear.

Transform Data into Knowledge 48 SPSS

This is the close dialogue box of Scatter/ Dot; where we select the Simple scatter

CGPA: is the dependent variable which we a studying

Transform Data into Knowledge 49 SPSS

Qunat: is the independent variable which relation on CGPA we are trying to find out

Select the independent variables as well as dependent variable. In this case independent variable is QNT and Dependent variable is CGPA.

Transform Data into Knowledge 50 SPSS

In this dialogue box, we shall click on Simple scatter Plot and then press OK. Now the following dialogue box will appear.

Now we shall press OK. Now the graph will appear in the SPSS statistics viewer as shown in the following diagram.

Transform Data into Knowledge 51 SPSS

The scatter plot appears in the out put view of the SPSS Shown in the following diagram,

Transform Data into Knowledge 52 SPSS

We shall double click on the graph; the following dialogue box will appear.

Transform Data into Knowledge 53 SPSS

To ad fit lineinthe graph this is the procedure

This is the linear line which does not truly represent the data i.e. is does not passes through maximum no of values. So is not the best method. We can improve this by applying a much better method because our data does not lies in linear symmetry.

Transform Data into Knowledge 54 SPSS

We go into the add fit line and select the LOESS curve this time,

Transform Data into Knowledge 55 SPSS

The Loess curve is shown in the graph; It passes through the more data rather the linear curve

The loess curve suggests that instead of fitting a linear curve, we should fit a quadratic curve. So we will put the quadratic curve; procedure is explained in the figure.

Transform Data into Knowledge 56 SPSS

Transform Data into Knowledge 57 SPSS

This graph shows the three curves .i.e. linear, loess and quadratic curve

Transform Data into Knowledge 58 SPSS

Now we will estimate that how quadratic curve is better from linear curve in this situation. Here we shall put the CGPA and QNT.

Transform Data into Knowledge 59 SPSS

heses are the steps I. Go to analyze II. Select the regression III. Select linear regression IV. Put the DV and IV V. Select the linear and quadratic VI. Select Display ANOVA VII. Press Ok This is the procedure of finding out the estimation curve and in the next steps on curves are shown rather the complete procedures

We also mark the option of Quadric and then mark display the anova table

Transform Data into Knowledge 60 SPSS

Now we shall press OK. Now the graph will appear in the SPSS statistics viewer as shown in the following diagram. Now we shall press OK. The graph will appear in the SPSS statistics viewer as shown in the following diagram.

Transform Data into Knowledge 61 SPSS

Model 1 Model Description Table: This table summarizes all the results of the model

Case Processing Company: It tells us about total cases, excluded cases, forecasted cases, newly created cases.

Variable Processing Summary: It tells us about number of positive values, number of zeroes, number of negative values, number of missing values of Dependent and Independent Variables.

Transform Data into Knowledge 62 SPSS

Linear Tables: CGPA of the student Linear Reporting Results = a+ b1x1 =2.726+.150x1 Interpretation When the QNT is0 then CGPA is 2.726. per unit increase in QNT, CGPA increase by .150 Intercept: This is the predicted value of the response variable i.e. CGPA performance when the predictor variables i.e. QNT is 0. In this case exam CGPA is 2.726, when the QNT is 0. The slope or the regression coefficient b1: b1 is equal to .150 which is the change in the response variable i.e. CGPA when the predictor variable i.QNT increases by one unit. To simplify our interpretation, we can say that with an increase in QNT by 1 unit, the performance would increase by 0.150.

Model Summary

Transform Data into Knowledge 63 SPSS

CGPA R Adjusted R Std. Error of of the Square the Estimate student Square .283 .265 .304

It tells us R, R square, Adjusted R Square and standard error of the estimates. If standard error of the estimates is less than the standard error of estimates of Quadratic tables than we dont apply the quadratic eq and vice versa. R square tells us how much variation in the dependent variable can be accounted for by the independent variable. R is the sample correlation coefficient between the dependent variable (sales during the year) and the independent variable. Standard error is measured in units of the response variable i.e. the sales during the year and it tells us the standard distance of the data values from the regression line or it tells us how far the values lie away from the regression line. For different model comparisons, we always look at the standard error. The model with smaller standard error will be a better model. The standard error in the modern summary suggests that we should fit a quadratic model instead of a linear model. For model comparison we can look at R2 only when both models have equal number of predictors. But in the current situation, in the linear model we have one predictor i.e QNT, and in the second model we have two predictors QNT and QNT square. So in this case instead of looking at the R square we shall be looking for model comparison standard error is the best fit model. The discussion of the quadratic model is beyond our syllabus.

ANOVA: ANOVA

Transform Data into Knowledge 64 SPSS

Sum of Squares Regressio n Residual Total 1.499 3.800 5.299

df 1 41 42

Mean Square

Sig. .000

1.499 16.173 .093

The independent variable is QNT. The ANOVA table shows us the overall impact of the model. It depicts the amount of variation in the response data explained by the predictor and the amount of variation left unexplained. We have the p-value which is the observed level of significance. If p< , then we reject Ho (significant) If p>, then we do not reject Ho (non-significant)

Coefficient: Coefficients Unstandardized Coefficients B QNT (Constant ) .150 2.726 Std. Error .037 .156 Standardize d Coefficients Beta .532 t 4.022 17.467 Sig. .000 .000

It tells us whether the predictor variables have a significant effect on model or not. If any predictor variable is not having significant impact then we exclude that variable which has not significant impact on our linear eq. Quadratic Tables:

Transform Data into Knowledge 65 SPSS

Quadratic = a+ b1x1+b2x1 2 =3.310-1.86 x1+ .043x12 Model Summary: Model Summary R .574 R Adjusted R Std. Error of Square Square the Estimate .330 .296 .298

The independent variable is QNT. It tells us R, R square, Adjusted R Square and standard error of the estimates. If standard error of the estimates is less than the standard error of estimates of Quadratic tables than we dont apply the quadratic eq and vice versa. R square tells us how much variation in the dependent variable can be accounted for by the independent variable. R is the sample correlation coefficient between the dependent variable (sales during the year) and the independent variable. Standard error is measured in units of the response variable i.e. the sales during the year and it tells us the standard distance of the data values from the regression line or it tells us how far the values lie away from the regression line. For different model comparisons, we always look at the standard error. The model with smaller standard error will be a better model. ANOVA: ANOVA Sum of Squares Regressio n Residual Total 1.748 3.551 5.299 df 2 40 42 Mean Square .874 .089 F 9.846 Sig. .000

The independent variable is QNT.

Transform Data into Knowledge 66 SPSS

The ANOVA table shows us the overall impact of the model. It depicts the amount of variation in the response data explained by the predictor and the amount of variation left unexplained. We have the p-value which is the observed level of significance. If p< , then we reject Ho (significant) If p>, then we do not reject Ho (non-significant) Coefficients: Coefficients Unstandardized Coefficients B QNT QNT ** 2 (Constant ) -.186 .043 3.310 Std. Error .204 .026 .381 Standardize d Coefficients Beta -.661 1.213 t -.914 1.675 8.697 Sig. .366 .102 .000

It tells us whether the predictor variables have a significant effect on model or not. If any predictor variable is not having significant impact then we exclude that variable which has not significant impact on our linear eq.

Transform Data into Knowledge 67 SPSS

Here we have graph in which observed, linear and quadratic values are shown.

Transform Data into Knowledge 68 SPSS

Model 2

Transform Data into Knowledge 69 SPSS

Again the previous procedures are carried

This is the simple graph having linear curve

Transform Data into Knowledge 70 SPSS

This is final shape of the graph after having done the complete procedure to apply the curves.

Transform Data into Knowledge 71 SPSS

Now the following dialogue box will appear and we shall interpret each dialogue box.

Transform Data into Knowledge 72 SPSS

Model 2 (CGPA and Verbal) Model Description Table: This table summarizes all the results of the model Model Description Model Name Dependent Variable Equation 1 1 2 Independent Variable Constant Variable Whose Values Label Observations in Plots Tolerance for Entering Terms in Equations MOD_1 CGPA of the student Linear Quadratic Verbal Included Unspecified .0001

Case Processing Company: It tells us about total cases, excluded cases, forecasted cases, newly created cases Case Processing Summary N Total Cases Excluded Casesa Forecasted Cases Newly Created Cases 43 1 0 0

a. Cases with a missing value in any variable are excluded from the analysis.

Transform Data into Knowledge 73 SPSS

Variable Processing Summary: It tells us about number of positive values, number of zeroes, number of negative values, number of missing values of Dependent and Independent Variables. Variable Processing Summary Variables Independe Dependent nt CGPA of the student Number of Positive Values Number of Zeros Number of Negative Values Number of Missing User-Missing Values System-Missing 43 0 0 0 0 Verbal 42 0 0 0 1

Linear Tables: CGPA of the student Linear Reporting Results = a+ b2x2 =2.366+.206x2 Interpretation When the Verbal is0 then CGPA is 2.366. Per unit increase in Verbal, CGPA increase by .206 Intercept: This is the predicted value of the response variable i.e. CGPA when the predictor variables i.e. Verbal is 0. In this case CGPA is 2.366, when the Verbal is 0.

Transform Data into Knowledge 74 SPSS

The slope or the regression coefficient b1: B2 is equal to .206 which is the change in the response variable i.e. CGPA when the predictor variable i.Verbal increases by one unit. To simplify our interpretation, we can say that with an increase in Verbal by 1 unit, the performance would increase by 0.206. Model Summary R .440 R Adjusted R Std. Error of Square Square the Estimate .194 .174 .316

The independent variable is Verbal. It tells us R, R square, Adjusted R Square and standard error of the estimates. If standard error of the estimates is less than the standard error of estimates of Quadratic tables than we dont apply the quadratic eq and vice versa. R square tells us how much variation in the dependent variable can be accounted for by the independent variable. R is the sample correlation coefficient between the dependent variable (sales during the year) and the independent variable. Standard error is measured in units of the response variable i.e. the sales during the year and it tells us the standard distance of the data values from the regression line or it tells us how far the values lie away from the regression line. For different model comparisons, we always look at the standard error. The model with smaller standard error will be a better model.

Transform Data into Knowledge 75 SPSS

ANOVA Sum of Squares Regressio n Residual Total .962 3.998 4.960 df 1 40 41 Mean Square .962 .100 F 9.625 Sig. .004

The independent variable is Verbal. The ANOVA table shows us the overall impact of the model. It depicts the amount of variation in the response data explained by the predictor and the amount of variation left unexplained. We have the p-value which is the observed level of significance. If p< , then we reject Ho (significant) If p>, then we do not reject Ho (non-significant)

Coefficient: Coefficients Unstandardized Coefficients B Verbal (Constant ) .206 2.366 Std. Error .066 .318 Standardize d Coefficients Beta .440 t 3.102 7.451 Sig. .004 .000

It tells us whether the predictor variables have a significant effect on model or not. If any predictor variable is not having significant impact then we exclude that variable which has not significant impact on our linear eq. There is no need to apply the quadratic eq to this model because standard error is greater than the standard error of linear

Transform Data into Knowledge 76 SPSS

Model 3 If we carry out the previous procedures than this again will be the result of model 3 where the variables are CGPA and Medium of instruction up to matric.

(CGPA and MOIM) Curve Fit [DataSet1] F:\Ijaz Bajwa. Data.sav

Warnings The Quadratic model could not be fitted due to near-collinearity among model terms.

Transform Data into Knowledge 77 SPSS

Model Description Table: This table summarizes all the results of the model Model Description Model Name Dependent Variable Equation 1 1 MOD_4 CGPA of the student Linear Medium of Institution upto Matric Included Unspecified

Independent Variable Constant Variable Whose Values Label Observations in Plots

Case Processing Company: It tells us about total cases, excluded cases, forecasted cases, newly created cases

Case Processing Summary N Total Cases Excluded Casesa Forecasted Cases Newly Created Cases 43 0 0 0

a. Cases with a missing value in any variable are excluded from the analysis.

Transform Data into Knowledge 78 SPSS

Variable Processing Summary: It tells us about number of positive values, number of zeroes, number of negative values, number of missing values of Dependent and Independent Variables.

Variable Processing Summary Variables Dependent Independent Medium of CGPA of Institution the student upto Matric Number of Positive Values Number of Zeros Number of Negative Values Number of Missing User-Missing Values System-Missing 43 0 0 0 0 27 16 0 0 0

CGPA of the student Linear Reporting Results = a+ b3x3 =3.218+.178x3 Interpretation When the MOIM is 0 then CGPA is 3.218. Per unit increase in MOIM, CGPA increase by .178 Intercept: This is the predicted value of the response variable i.e. CGPA when the predictor variables i.e. MOIM is 0. In this case CGPA is 3.218, when the MOIM is 0.

Transform Data into Knowledge 79 SPSS

The slope or the regression coefficient b1: B3 is equal to .178 which is the change in the response variable i.e. CGPA when the predictor variable i.e MOIM increases by one unit. To simplify our interpretation, we can say that with an increase in MOIM by 1 unit, the CGPA would increase by 0.178. Model Summary Model Summary R .237 R Adjusted R Std. Error of Square Square the Estimate .056 .033 .349

The independent variable is Medium of Institution upto Matric.

It tells us R, R square, Adjusted R Square and standard error of the estimates. If standard error of the estimates is less than the standard error of estimates of Quadratic tables than we dont apply the quadratic eq and vice versa. R square tells us how much variation in the dependent variable can be accounted for by the independent variable. R is the sample correlation coefficient between the dependent variable (sales during the year) and the independent variable. Standard error is measured in units of the response variable i.e. the sales during the year and it tells us the standard distance of the data values from the regression line or it tells us how far the values lie away from the regression line. For different model comparisons, we always look at the standard error. The model with smaller standard error will be a better model.

Transform Data into Knowledge 80 SPSS

ANOVA ANOVA Sum of Squares Regressio n Residual Total .298 5.002 5.299 df 1 41 42 Mean Square .298 .122 F 2.440 Sig. .126

The independent variable is Medium of Institution upto Matric.

The ANOVA table shows us the overall impact of the model. It depicts the amount of variation in the response data explained by the predictor and the amount of variation left unexplained. We have the p-value which is the observed level of significance. If p< , then we reject Ho (significant) If p>, then we do not reject Ho (non-significant) Coefficient Coefficients Unstandardized Coefficients B Medium of Institution upto Matric (Constant) .172 Std. Error .110 Standardize d Coefficients Beta .237 t 1.562 Sig. .126

3.218

.087

36.847

.000

It tells us whether the predictor variables have a significant effect on model or not. If any predictor variable is not having significant impact then we exclude that variable which has not significant impact on our linear eq. There is no need to apply the quadratic eq to this model because standard error is greater than the standard error of linear.

Transform Data into Knowledge 81 SPSS

When we calculate quadratic the Se becomes same so we dont need to apply the quadratic eq.

Transform Data into Knowledge 82 SPSS

Model 4 (CGPA and Inter pct)

Transform Data into Knowledge 83 SPSS

Transform Data into Knowledge 84 SPSS

Model Description Table: This table summarizes all the results of the model (CGPA and Inter pct) Model Description Model Name Dependent Variable Equation 1 1 2 Independent Variable MOD_6 CGPA of the student Linear Quadratic Intermediate Percentage of the student Included Unspecified .0001

Constant Variable Whose Values Label Observations in Plots Tolerance for Entering Terms in Equations

Case Processing Company: It tells us about total cases, excluded cases, forecasted cases, newly created cases Case Processing Summary N Total Cases Excluded Cases Newly Created Cases
a

43 0 0 0

Forecasted Cases

a. Cases with a missing value in any variable are excluded from the analysis.

Transform Data into Knowledge 85 SPSS

Variable Processing Summary: It tells us about number of positive values, number of zeroes, number of negative values, number of missing values of Dependent and Independent Variables.

Variable Processing Summary Variables Dependent Independent Intermediat e Percentage CGPA of of the the student student Number of Positive Values Number of Zeros Number of Negative Values Number of Missing User-Missing Values System-Missing 43 0 0 0 0 43 0 0 0 0

CGPA of the student Linear = a+ b4x4 =3.052+.004x4 Interpretation When the inter pct is 0 then CGPA is 3.052. Per unit increase in inter pct, CGPA increase by .178 Intercept: This is the predicted value of the response variable i.e. CGPA when the predictor variables i.e. Inter pct is 0. In this case CGPA is 3.0525, when the inter pct is 0.

Transform Data into Knowledge 86 SPSS

The slope or the regression coefficient b1: B4 is equal to .004 which is the change in the response variable i.e. CGPA when the predictor variable i.e inter pct increases by one unit. To simplify our interpretation, we can say that with an increase in inter pct by 1 unit, the CGPA would increase by 0.004 Model Summary Model Summary R .086 R Adjusted R Std. Error of Square Square the Estimate .007 -.017 .358

The independent variable is Intermediate Percentage of the student. It tells us R, R square, Adjusted R Square and standard error of the estimates. If standard error of the estimates is less than the standard error of estimates of Quadratic tables than we dont apply the quadratic eq and vice versa. R square tells us how much variation in the dependent variable can be accounted for by the independent variable. R is the sample correlation coefficient between the dependent variable (sales during the year) and the independent variable. Standard error is measured in units of the response variable i.e. the sales during the year and it tells us the standard distance of the data values from the regression line or it tells us how far the values lie away from the regression line. For different model comparisons, we always look at the standard error. The model with smaller standard error will be a better model.

Transform Data into Knowledge 87 SPSS

ANOVA Sum of Squares Regressio n Residual Total .040 5.260 5.299 df 1 41 42 Mean Square .040 .128 F .309 Sig. .581

The independent variable is Intermediate Percentage of the student. The ANOVA table shows us the overall impact of the model. It depicts the amount of variation in the response data explained by the predictor and the amount of variation left unexplained. We have the p-value which is the observed level of significance. If p< , then we reject Ho (significant) If p>, then we do not reject Ho (non-significant) Coefficient Coefficients Unstandardized Coefficients B Intermediate Percentage of the student (Constant) .004 Std. Error .007 Standardize d Coefficients Beta .086 t .556 Sig. .581

3.052

.496

6.151

.000

It tells us whether the predictor variables have a significant effect on model or not. If any predictor variable is not having significant impact then we exclude that variable which has not significant impact on our linear eq. There is no need to apply the quadratic eq to this model because standard error is greater than the standard error of linear.

Transform Data into Knowledge 88 SPSS

There is no need to apply the quadratic eq to this model because standard error is greater than the standard error of linear.

Model 5 (CGPA and Institution from which inter done)

Transform Data into Knowledge 89 SPSS

Model Description Table: This table summarizes all the results of the model (CGPA and Institution from which inter done) Model Description Model Name Dependent Variable Equation 1 1 2 Independent Variable Constant Variable Whose Values Label Observations in Plots Tolerance for Entering Terms in Equations MOD_8 CGPA of the student Linear Quadratic Institution from which interdone Included Unspecified .0001

Case Processing Company: It tells us about total cases, excluded cases, forecasted cases, newly created cases Case Processing Summary N Total Cases Excluded Casesa Forecasted Cases Newly Created Cases 43 0 0 0

Transform Data into Knowledge 90 SPSS

Variable Processing Summary: It tells us about number of positive values, number of zeroes, number of negative values, number of missing values of Dependent and Independent Variables. Variable Processing Summary Variables Dependent Independent Institution CGPA of from which the student interdone Number of Positive Values Number of Zeros Number of Negative Values Number of Missing User-Missing Values System-Missing 43 0 0 0 0 42 1 0 0 0

CGPA of the student Linear = a+ b5x5 =3.379-.38x4 Interpretation When the institution from which inter done is 0 then CGPA is 3.379. Per unit increase in institution from which inter done, CGPA decrease by .38 Intercept: This is the predicted value of the response variable i.e. CGPA when the predictor variables i.e. institution from which inter done is 0. In this case CGPA is 3.379, when the institution from which inter done is 0. The slope or the regression coefficient b1: B5 is equal to - .38 which is the change in the response variable i.e. CGPA when the predictor variable i.e institution from which inter done increases by one unit. To simplify our interpretation, we can say that with an increase in institution from which inter done by 1 unit, the CGPA would decrease by 0.38

Transform Data into Knowledge 91 SPSS

Model Summary Model Summary R .058 R Adjusted R Std. Error of Square Square the Estimate .003 -.021 .359

The independent variable is Institution from which interdone. It tells us R, R square, Adjusted R Square and standard error of the estimates. If standard error of the estimates is less than the standard error of estimates of Quadratic tables than we dont apply the quadratic eq and vice versa. R square tells us how much variation in the dependent variable can be accounted for by the independent variable. R is the sample correlation coefficient between the dependent variable (sales during the year) and the independent variable. Standard error is measured in units of the response variable i.e. the sales during the year and it tells us the standard distance of the data values from the regression line or it tells us how far the values lie away from the regression line. For different model comparisons, we always look at the standard error. The model with smaller standard error will be a better model.

ANOVA ANOVA Sum of Squares Regressio n Residual Total .018 5.282 5.299 df 1 41 42 Mean Square .018 .129 F .138 Sig. .712

The independent variable is Institution from which interdone.

Transform Data into Knowledge 92 SPSS

The ANOVA table shows us the overall impact of the model. It depicts the amount of variation in the response data explained by the predictor and the amount of variation left unexplained. We have the p-value which is the observed level of significance. If p< , then we reject Ho (significant) If p>, then we do not reject Ho (non-significant) Coefficient Coefficients Unstandardized Coefficients B Institution from which interdone (Constant) -.038 3.379 Std. Error .102 .154 Standardize d Coefficients Beta -.058 t -.371 21.907 Sig. .712 .000

It tells us whether the predictor variables have a significant effect on model or not. If any predictor variable is not having significant impact then we exclude that variable which has not significant impact on our linear eq. There is no need to apply the quadratic e.g. to this model because standard error is greater than the standard error of linear. There is no need to apply the quadratic e.g. to this model because standard error is greater than the standard error of linear.

Transform Data into Knowledge 93 SPSS

Transform Data into Knowledge 94 SPSS

Regression with more than one independent Variables (= b0+ b1x1+b2x2+b3x3+b4 x4+b5 x5)

We shall go in analyze, click on regression then click on linear. When we click on linear then following dialogue box will appear. Now we shall select the dependent variable which is CGPA and independent variable which are QNT, inter pct, Institution from which inter done, verbal, MOIM.

Transform Data into Knowledge 95 SPSS

Now we shall click on statistics and check the following icons.

Transform Data into Knowledge 96 SPSS

When we click ok than following tables appear in out put file.

Transform Data into Knowledge 97 SPSS

Descriptive Statistics: Regression [DataSet1] E:\Education\IjazData\Statistical inference\IjazData.sav

Descriptive Statistics Mean CGPA of the student QNT Medium of Institution upto Matric Intermediate Percentage of the student Institution from which interdone Verbal 3.3393 4.0119 .64 67.02 Std. Deviation .34783 1.27642 .485 7.592 N 42 42 42 42

1.43 4.7302

.547 .74440

42 42

In this we find the mean and SD.

Transform Data into Knowledge 98 SPSS

Variables Entered/Removed:

Variables Entered/Removed Model 1 Variables Entered Verbal, QNT, Institution from which interdone, Intermediate Percentage of the student, Medium of Institution upto Matrica Variables Removed Method . Enter

a. All requested variables entered. In this we can see which variables are entered and which are removed.

Model:
We can input more than one model in a same regression command in SPSS. And in this column the numbers of models are shown.

Variables Entered:
This column tells us about all the independent variable that we have specified but did not blocked as SPSS allows us to enter variables in block for stepwise regression.

Variables removed:
Usually, this column is empty and only lists the removed variables when we do stepwise regression.

Methods:
The method used by SPSS to run regression is mentioned in this column. Enter means that every independent variable was entered in usual manner. But when stepwise regression is done, the entry tells us about that.

Transform Data into Knowledge 99 SPSS

Model Summary:

Model Summary Change Statistics Mod el 1 Std. Error F R Adjusted of the R Square Chang Square R Square Estimate Change e .503 .434 .26171 .503 7.284 Sig. F Change .000

R .709a

df1 5

df2 36

a. Predictors: (Constant), Verbal, QNT, Institution from which interdone, Intermediate Percentage of the student, Medium of Institution upto Matric

Model:
This column tells us the number of models reported. As in SPSS we can input more than one model in a same regression command.

R:
The correlation between the observed and predicted values of dependent variable is called R and is the square root of R-Squared.

R-Square:
R-Square is the proportion of variance in the dependent variable (CGPA) which can be predicted from the independent variables (VER, QNT, Institution of Inter, Inter-percentage and MOIM). This value indicates that 50.3% of the variance in CGPA can be predicted from the variables VER, QNT, Institution of Inter, Inter-percentage and MOIM. Note that this is an overall measure of the strength of association, and does not reflect the extent to which any particular independent variable is associated with the dependent variable. RSquare is also called the coefficient of determination.

Adjusted R-square:
As predictors are added to the model, each predictor will explain some of the variance in the dependent variable simply due to chance. One could continue to add predictors to the model which would continue to improve the ability of the

Transform Data into Knowledge 100 SPSS

predictors to explain the dependent variable, although some of this increase in R-square would be simply due to chance variation in that particular sample. The adjusted R-square attempts to yield a more honest value to estimate the Rsquared for the population. The value of R-square was .503, while the value of Adjusted R-square was .434, Adjusted R-squared is computed using the formula 1 - ((1 - Rsq)(N - 1 )/ (N - k - 1)). From this formula, you can see that when the number of observations is small and the number of predictors is large, there will be a much greater difference between R-square and adjusted R-square (because the ratio of (N - 1) / (N - k - 1) will be much greater than 1). By contrast, when the number of observations is very large compared to the number of predictors, the value of R-square and adjusted R-square will be much closer because the ratio of (N - 1)/(N - k - 1) will approach 1.

Std. Error of the Estimate:


The standard error of the estimate, also called the root mean square error, is the standard deviation of the error term, and is the square root of the Mean Square Residual (or Error). ANOVA

ANOVAb Model 1 Regressio n Residual Total Sum of Squares 2.495 2.466 4.960 df 5 36 41 Mean Square .499 .068 F 7.284 Sig. .000a

a. Predictors: (Constant), Verbal, QNT, Institution from which interdone, Intermediate Percentage of the student, Medium of Institution upto Matric b. Dependent Variable: CGPA of the student

Transform Data into Knowledge 101 SPSS

The ANOVA table shows us the overall impact of the model. It depicts the amount of variation in the response data explained by the predictor and the amount of variation left unexplained. We have the p-value which is the observed level of significance. If p< , then we reject Ho (significant) If p>, then we do not reject Ho (non-significant)

Model:
This column tells us the number of models reported. As in SPSS we can input more than one model in a same regression command. This is the source of variance, Regression, Residual and Total. The Total variance is partitioned into the variance which can be explained by the independent variables (Regression) and the variance which is not explained by the independent variables (Residual, sometimes called Error). Note that the Sums of Squares for the Regression and Residual add up to the Total, reflecting the fact that the Total is partitioned into Regression and Residual variance.

Sum of Squares:
These are the Sum of Squares associated with the three sources of variance, Total, Model and Residual. These can be computed in many ways. Conceptually, these formulas can be expressed as: SSTotal The total variability around the mean. S(Y - Ybar)2. SSResidual The sum of squared errors in prediction. S(Y - Ypredicted)2. SSRegression The improvement in prediction by using the predicted value of Y over just using the mean of Y. Hence, this would be the squared differences between the predicted value of Y and the mean of Y, S(Ypredicted Ybar)2. Another way to think of this is the SSRegression is SSTotal SSResidual. Note that the SSTotal = SSRegression + SSResidual. Note that SSRegression / SSTotal is equal to 4.960, the value of R-Square. This is because R-Square is the proportion of the variance explained by the independent variables, hence can be computed by SSRegression / SSTotal.

Df:
These are the degrees of freedom associated with the sources of variance. The total variance has N-1 degrees of freedom. In this case, there were N=43

Transform Data into Knowledge 102 SPSS

students, so the DF for total is 42. The model degrees of freedom corresponds to the number of predictors minus 1 (K-1). You may think this would be 5-1 (since there were 5 independent variables in the model, VER, QNT, Institution of Inter, Inter-percentage and MOIM). But, the intercept is automatically included in the model (unless you explicitly omit the intercept). Including the intercept, there are 6 predictors, so the model has 6-1=5 degrees of freedom. The Residual degrees of freedom is the DF total minus the DF model, 42 - 5 is 37.

Mean Square:
These are the Mean Squares; the Sum of Squares divided by their respective DF. For the Regression, 2.495/ 5 = .499 . For the Residual, 2.466/ 36= .068 . These are computed so you can compute the F ratio, dividing the Mean Square Regression by the Mean Square Residual to test the significance of the predictors in the model.

F and Sig.:
The F-value is the Mean Square Regression (.499) divided by the Mean Square Residual (0.068), yielding F=7.284. The p-value associated with this F value is very small (0.000). These values are used to answer the question "Do the independent variables reliably predict the dependent variable?". The p-value is compared to your alpha level (typically 0.05) and, if smaller, you can conclude "Yes, the independent variables reliably predict the dependent variable". You could say that the group of variables VER, QNT, Institution of Inter, Interpercentage and MOIM can be used to reliably predict CGPA (the dependent variable). If the p-value were greater than 0.05, you would say that the group of independent variables does not show a statistically significant relationship with the dependent variable, or that the group of independent variables does not reliably predict the dependent variable. Note that this is an overall significance test assessing whether the group of independent variables when used together reliably predict the dependent variable, and does not address the ability of any of the particular independent variables to predict the dependent variable. The ability of each individual independent variable to predict the dependent variable is addressed in the table below where each of the individual variables are listed.

Transform Data into Knowledge 103 SPSS

Coefficients:

Coefficientsa Standardi zed Coefficie nts Beta t 3.249 .567 4.349 .131 .931 Sig. .003 .000 .358

Unstandardized Coefficients Model 1 (Constant) QNT Medium of Institution upto Matric Intermediate Percentage of the student Institution from which interdone Verbal B 1.623 .155 .094 Std. Error .500 .036 .101

95.0% Confidence Interval for B Lower Bound .610 .082 -.111 Upper Bound 2.637 .227 .299

.002

.006

.046

.348

.730

-.010

.014

.025 .182

.080 .063

.040

.312

.757 .006

-.138 .055

.188 .309

.389 2.902

a. Dependent Variable: CGPA of the student It tells us whether the predictor variables have a significant effect on model or not. If any predictor variable is not having significant impact then we exclude that variable which has not significant impact on our linear eq.

Model:
This column tells us the number of models reported. As in SPSS we can input more than one model in a same regression command. This column shows the predictor variables (constant, VER, QNT, Institution of Inter, Inter-percentage and MOIM). The first variable (constant) represents the constant, also referred to in textbooks as the Y intercept, the

Transform Data into Knowledge 104 SPSS

height of the regression line when it crosses the Y axis. In other words, this is the predicted value of CGPA when all other variables are 0.

B:
These are the values for the regression equation for predicting the dependent variable from the independent variable. These are called unstandardized coefficients because they are measured in their natural units. As such, the coefficients cannot be compared with one another to determine which one is more influential in the model, because they can be measured on different scales. For example, how can you compare the inter-percentage with the CGPA scores? The regression equation can be presented in many different ways, for example:

Y -Predicted = b0 + b1*x1 + b2*x2 + b3*x3 + b3*x3 + b4*x4+b5*x5


The column of estimates (coefficients or parameter estimates, from here on labeled coefficients) provides the values for b0, b1, b2, b3 and b4 for this equation. Expressed in terms of the variables used in this example, the regression equation is

CGPA = 1.623+ .155*QNT +.182*VER +.094*MOIM+.002*interpercent+.025* inter-institution


These estimates tell you about the relationship between the independent variables and the dependent variable. These estimates tell the amount of increase in science scores that would be predicted by a 1 unit increase in the predictor. Note: For the independent variables which are not significant, the coefficients are not significantly different from 0, which should be taken into account when interpreting the coefficients. (See the columns with the t-value and p-value about testing whether the coefficients are significant).

Std. Error:
These are the standard errors associated with the coefficients. The standard error is used for testing whether the parameter is significantly different from 0 by dividing the parameter estimate by the standard error to obtain a t-value (see

Transform Data into Knowledge 105 SPSS

the column with t-values and p-values). The standard errors can also be used to form a confidence interval for the parameter, as shown in the last two columns of this table.

Beta:
These are the standardized coefficients. These are the coefficients that you would obtain if you standardized all of the variables in the regression, including the dependent and all of the independent variables, and ran the regression. By standardizing the variables before running the regression, you have put all of the variables on the same scale, and you can compare the magnitude of the coefficients to see which one has more of an effect. You will also notice that the larger betas are associated with the larger t-values.

T and Sig.:
These columns provide the t-value and 2 tailed p-value used in testing the null hypothesis that the coefficient/parameter is 0. If you use a 2 tailed test, then you would compare each p-value to your preselected value of alpha. Coefficients having p-values less than alpha are statistically significant. For example, if you chose alpha to be 0.05, coefficients having a p-value of 0.05 or less would be statistically significant (i.e., you can reject the null hypothesis and say that the coefficient is significantly different from 0). If you use a 1 tailed test (i.e., you predict that the parameter will go in a particular direction), then you can divide the p-value by 2 before comparing it to your preselected alpha level. However, if you used a 2-tailed test and alpha of 0.01, the p-value of .0255 is greater than 0.01 and the coefficient for variable would not be significant at the 0.01 level. Had you predicted that this coefficient would be positive (i.e., a one tail test), you would be able to divide the p-value by 2 before comparing it to alpha. This would yield a one-tailed p-value of 0.00945, which is less than 0.01 and then you could conclude that this coefficient is greater than 0 with a one tailed alpha of 0.01. The constant is significantly different from 0 at the 0.05 alpha levels. However, having a significant intercept is seldom interesting.

Transform Data into Knowledge 106 SPSS

The coefficient for QNT (.155) is statistically significantly different from 0 using alpha of 0.05 because its p-value is 0.000, which is smaller than 0.05. The coefficient for MOIM (.094) is not statistically significantly different from 0 because its p-value is definitely larger than 0.05. The coefficient for VER (.182) is statistically significant because its p-value of 0.008 is less than .05. The coefficient for INTER.Inst (.025) is not significantly different form 0 because its p-value is greatly larger than 0.05. The coefficient for INTER-Percent (.002) is not significantly different from 0 because its value is definitely larger than 0.05.

Reporting Results (Combine)


= b0+ b1x1+b2x2+b3x3+b4 x4+b5 x5 =1.623+.155 x1+ .182x2+.094x3+ .002x4+ .025x5 Interpretation: When all the independent variables are 0 then the CGPA is 1.623. Per unit increase in QNT, the value of CGPA increase by .155; Per unit increase in verbal, CGPA increase by .182; Per unit increase in MOIM, CGPA increase by .094; Per unit increase in inter pct, CGPA increase by .002; Per unit increase in institute from which inter done, CGPA increase by .025. Intercept: This is the predicted value of the response variable i.e. CGPA when the predictor variables is 0. In this case CGPA is 1.623, when the all the predictors are 0. The slope or the regression coefficient: When all the independent variables are 0 then the CGPA is 1.623. Per unit increase in QNT, the value of CGPA increase by .155; Per unit increase in verbal, CGPA increase by .182; Per unit increase in MOIM, CGPA increase by .094; Per unit increase in inter pct, CGPA increase by .002; Per unit increase in institute from which inter done, CGPA increase by .025.

Transform Data into Knowledge 107 SPSS

Transform Data into Knowledge 108 SPSS

Transform Data into Knowledge 109 SPSS

Transform Data into Knowledge 110 SPSS

Transform Data into Knowledge 111 SPSS

Transform Data into Knowledge 112 SPSS

Transform Data into Knowledge 113 SPSS