Name:Iltaf hussain
Class roll No:96
Reg:5443
Submitted to: Prof Adnan sir
Department: Staictics
Semester:5 th
Govt Postgraduate College Charsadda
Topic :
binary logisctic
regression
DATA VIEW
VARIABLE VIEW
Procedure:
First I entered the data in variable view and then
put vale in data view then go to analyze menu bar
then click on regression and then click on binary
logistic
Lung cancer in depndent other in covariates then go to
option select 95%CI
RESULT
Interpretation:
• Here’s a comprehensive summary of the logistic regression output you provided Summary of Logistic Regression
AnalysisThe logistic regression model examines the relationship between several predictors (Smoking, Age, Gender,
Monthly Income, and Education) and the likelihood of lung cancer. Below is an interpretation of the findings:---1. Model
PerformanceClassification Accuracy:The model correctly classified 85.0% of cases overall.It achieved 75.0% accuracy for
predicting "NO" (no lung cancer) and 91.7% accuracy for predicting "YES" (lung cancer).Cut Value: The model uses a
probability threshold of 0.5 to classify observations as "YES" or "NO" for lung cancer.---2. Variables in the EquationThe
predictors in the logistic regression model and their influence on lung cancer are as follows:SmokingEffect: Smokers are
2.674 times more likely to have lung cancer compared to non-smokers.Significance: The effect of smoking is not
statistically significant (p = 0.485), meaning the evidence is insufficient to confirm a reliable relationship in this
sample.AgeEffect: Older age slightly reduces the odds of lung cancer, with the odds decreasing by 4.6% per year of
age.Significance: The relationship is not statistically significant (p = 0.687).GenderEffect: Gender has a significant impact
on lung cancer. Men are much less likely to have lung cancer compared to women, with their odds reduced by
approximately 98% (Exp(B) = 0.022).Significance: This variable is statistically significant (p = 0.046).Monthly IncomeEffect:
Monthly income has no meaningful impact on the likelihood of lung cancer, as the odds remain unchanged (Exp(B) =
1.000).Significance: This variable is not statistically significant (p = 0.559).EducationEffect: Higher education reduces the
odds of lung cancer by about 85% (Exp(B) = 0.153).Significance: This effect is not statistically significant (p =
0.193).Constant TermThe constant represents the baseline log-odds of lung cancer when all predictors are zero. It is not
statistically significant (p = 0.197).---3. Key FindingsGender is the only variable that shows a statistically significant
relationship with lung cancer (p = 0.046). Men are substantially less likely to develop lung cancer compared to
women.While variables like smoking and education have meaningful odds ratios (2.674 and 0.153, respectively), their
relationships are not statistically significant in this sample.Other factors, such as age and income, show negligible or no
impact on lung cancer likelihood.---4. ImplicationsThe model demonstrates good classification performance with an
overall accuracy of 85.0%. However, the limited statistical significance of most predictors suggests the model may benefit
from a larger sample size or additional variables to better understand the factors influencing lung cancer.Gender stands
out as a significant predictor and warrants further investigation into gender-specific risk factors for lung cancer.---Let me
know if you need further refinements or explanations!
