I would like to choose a different population. The population I would choose is the weight of newborn babies in smoking and non-smoking moms. The numerical variable that I would like to use is weight. I would like to calculate both the probability of a smoking/non-smoking woman having low birth weight babies or normal weight babies. The data will be quantitative and numerical because it will include numbers.

The weight of the babies can be represented by normal distribution. One reason is because babies tend to be around similar weights. Also if we look at the data on a histogram of weights, I would speculate that we would come up with one mode and not bimodal. I am pretty sure that there will be exceptions in weight as far as min and max, but for the most part the weight of babies should be very similar overall.

The features should be normal to normal distribution as far as the weights being similar. I believe that the data will be symmetrical, the features that may differ at times may be the weights of babies that are extremely low weight or extremely heavy which would generate outliers on a data representation.

In your response, identify another variable you might be interested in studying in their hypothetical population that can be modeled by a random variable with a named distribution - there are a few the book covers (Bernoulli, Binomial, Geometric, normal, to name a few). There are others we haven't covered but you can look up on wikipediaLinks to an external site.. Explain why you believe this variable might be appropriately modeled by this probability distribution. If the distribution is new, please give a brief overview of its features, and cite any sources you quote.