Logistic Regression Example using Minitab

 A manufacturer of mobile phones is interested in studying the durability of the device when dropped. A drop test study was performed dropping phones at various heights (from 1 to 8 feet in height) and observing the number tested that experience some type of fracture of the screen. Two material types were part of the study (A and B with B being the more expensive and hopefully more durable material - 50% more expensive to manufacture). Using the data in the .xls file FRACTURE TEST, build a logistic regression model (for each material type) that predicts the proportion of the screens expected to fracture as a function of the height of the drop. Use the results to describe the durability of the screens comparing the two materials. Examples of some questions to address are: Is Material B substantially more durable than A? Do you think it justifies the additional Manufacturing cost? What height do your models predict 25%, 50%, 75%, and 100% of the screens will fracture for each material type? What proportion of the screens do you predict will fracture at a height of 5 feet?

HINT: the Easiest way to do a Binary logistic regression Analysis with one variable using Minitab is through
Stat > Regression > Binary Fitted Line Plot.

Minitab Problem Data

Design	Height	Tested	Fractured
A	1	50	0
A	2	50	2
A	3	50	12
A	4	50	26
A	5	50	43
A	6	50	50
A	7	50	49
A	8	50	50
B	1	50	0
B	2	50	0
B	3	50	7
B	4	50	16
B	5	50	31
B	6	50	44
B	7	50	48
B	8	50	50

NB: 

For a step by step explanation on how to input the data in Minitab and input the variables in the dialogue boxes, email statistics experts or send an email to statistics answered website.