Blog Post 1

Our project will explore baby names in the United States, based on geographic and time trends from Social Security records, as well as sentiment based on ratings from BehindTheName.com.

We've done a bit of exploratory analysis so far on individual data sets.

We have studied where the 10 most popular names for each year appear in each state. Some of the data was quite unsurprising (Austin first appeared in Texas), and we plan to explore what geographic features correlate with a name becoming popular. We will look for correlation between likelihood of a popular name coming from a state, and that state's demographics, population, number of large cities, etc.

We also plan to build a name exploration tool, where you will be able to find name whose sentiment most closely matches yours.

We have also analyzed trends in some of the most popular female and male names, using the national data from the Social Security records. Below are the graphs for the trends of any name that has been #1 in popularity in the US.

Visualizing this data led to some interesting observations about the trends. For example, looking at the male names graph, the names Robert and James have very similar trend lines for the entire time period that we have access to. In addition, Michael and David also have a similar trend line up until about 1960, but then David's trend line matches those of Robert and James after that year. In the future, we will continue to explore these similarities.

We're curious to see if the 'sentiment' of a name (from ratings) can help us predict its popularity. For example, maybe the names Robert and James have very similar sentiments, which is why their popularity over time has fluctuated in similar ways. Or, maybe names with a close to 50/50 Masculine-Feminine rating have become more popular in the past decade.