Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 10

Dataset #1

Bank Preference
90/341
• 26% preferred public banks
251/341
• 74% preferred public banks
ln(odds) = 1.026 predicted odds
• how likely it is that [Exp(B)] = 2.789
something particular will The predicted odds of
happen deciding to continue
• Focus on the sign of the B the research is 2.789

Intercept-Only Model
Omnibus Test of Model Coefficients

Goodness-of-fit statistics help you to determine


whether the model adequately describes the data.

Added the Predictor Variables


Hosmer and Lemeshow Test

A Hosmer &
Lemeshow Test value
of .790 (unsignificant)
of the logistic
regression model
indicates a good fit of
data
An estimate of around 16.5% proportion variance of the
dependent variable associated with the predictor
variables is explainable by the model
SENSITIVITY of Prediction: 76%
Ln(odds) = 0.612 + 0.318VAS – 0.512PR + 0.221Reputation – 0.245PerceivedCost
• A binary logistic regression was performed to ascertain the effects of gender, employment,
technology, interest rates, value-added services, perceived risk, reputation, attractiveness,
and perceived cost on the likelihood of participants’ bank preference. The logistic
regression model was statistically significant, with Chi-squared (χ2)= 40.7, p = .000 (<0.05).
The model explained 16.5% (Nagelkerke R2) of the variance in bank preference and
correctly classified 76% of cases.

• The odds of a customer choosing Private Bank offering Value Added Services are 1.374
times higher than those Public Sector Banks that do not offer Value Added Services

• An increased level in the perceived risks decreases the odds of a customer choosing a
Private Bank by 0.6

• For every level of the bank’s reputation, the odds of a customer choosing a Private Bank is
1.248 times (25%) higher than public sector banks.

• Higher perceived cost decreases the odds of choosing Private Banks by 0.782.

You might also like