Click here to jump right to the ranking >>
The AreaVibes ranking system, like any complex metric, has a large element of subjectivity. Someone decided what variables to use, how to score them, and how to weigh them to get the overall composite livability index.
Although each of the component numbers is objective, the way in which the numbers are put together and used is not. Looking at the 88
The AreaVibes ranking seems much better suited to someone a bit further along in their life development than I am. I'm a somewhat reckless, single 20-something trying to start a career. Buying a house, settling down, and worrying about what school district I live in are completely alien concepts. So, I built the Shiftless Millennial City Index (SMCI) to better help me figure out where I should live.
Here are the factors that went into the SMCI:
Parks - A thriving park system is essential to a livable city. As much as I love urban living, I do need, from time to time, to reconnect with nature, or at least a heavily curated version of nature. Parks are also emblematic of a larger community spirit. Parks are places where people go together to relax, congregate, and just enjoy life. 15% of the SMCI consists of the percentage of a city's land area dedicated to parkland, gathered from The Trust for Public Land's 2012 "City Park Facts" report.[i]
Music/Event Venues - Shiftless millennials love going to shows. That means a city with a large number of venues, relative to its population, for concerts and other events is ideal. To get a rough idea of the number of large venues near a city, I looked for venues listed on TicketMaster.com for each of the studied cities, and adjusted the number based on the population of the city. This component makes up 10% of the SMCI.
Bars - Shiftless millennials also love drinking in groups. Bars per capita is thus a component of the SMCI. I used a similar protocol as with event venues - I went to Yelp.com, searched for bars near each of the studied cities, and noted the number of results. Bars per capita makes up 15% of the SMCI.
Single People - Now that we've identified some places to go and things to do, we need to find people with whom we can enjoy these things. As a measure of the prevalence of single folks in a city, we take the percentage of the population that's never been married, courtesy of the Missouri Census Data Center. As finding good people to mingle with is a very important goal for the shiftless millennial, this component accounts for 25% of the SMCI.
Public Transportation - To get to our parks and venues and dates, shiftless millennials need a way to get around. As a generation, we are far fonder of public transit and walking, and less enthusiastic about driving, than our parents and grandparents. For a snapshot of the transit use in a city, we use the percentage of the population that commutes using public transportation, again coming from the Missouri Census Data Center, and this measure gives us 15% of the SMCI.
Cost of Living - Shiftless millennials are often poor, and so cost of living is the final component of the SMCI. I gathered CoL index data for each city from real estate analysis site Sperling's Best Places (www.bestplaces.net). Cost of Living makes up 20% of the SMCI.
To determine the SMCI for each city, we first take each of the above measures and turn it into an index by taking each city's value, dividing by the average value for that measure, and then multiplying by 100[ii]. For example, Mesa, AZ has about 29 Yelp-rated bars per 10,000 residents, compared to the average among the 88 cities of about 22.3 bars per 10,000 residents, giving Mesa a Yelp Bar Index of (29/22.3)*100 = 130.
This index is handy for a couple reasons. First, it nicely represents each measure as a percentage of the average, so Mesa has 30% more bars per 10,000 residents than the average city in this study. Second, this index is "unit-less," and all of the indexes are on the same scale, so we can nicely combine our different components into the overall SMCI - we don't have to worry about what it means to add bars per capita to the percent of a city's land area dedicated to parks.
To get the full SMCI, we take a weighted average of the component indexes, using the weights outlined above. We then ranked the cities based on their SMCI's, and plotted our rankings against the AreaVibes rankings:
Andy Kiersz/Business Insider
The further to the left a city is, the better its AreaVibes ranking. The further down a city is, the better its Shiftless Millennial ranking. So, for example, Fremont, CA did very well on both ranking systems - Areavibes ranked it 5th, and it's the 8th best city for a shiftless millennial.
The most interesting feature of this graph are the two large, semi-coherent blobs of cities in the upper left and lower right corners. The upper left cities are cities that scored well on the AreaVibes ranking, but poorly on the SMCI ranking. Many of these cities are sunbelt enclaves. The lower right cities are those that I find underrated by AreaVibes - they had low AreaVibes scores, but ranked very well with my metric. These are predominantly northern and coastal cities, which I much prefer over the South, so my metric did a very nice job of assessing cities I would like to live in.
The underrated cities are those far down and to the right on the chart, with the largest differences between AreaVibes rankings and SMCI rankings.
[i] Three of the cities - Montgomery, AL; Shreveport, LA; and Akron, OH, were not included in the TPL report. For these cities, the park index is omitted, and the weights of the other components are adjusted accordingly.
[ii] Cost of Living is an exception here - first, I used a measure that was already an index, so no division was necessary. Second, and more importantly, for all of our other components, the higher the index, the better the city. A higher park index means more parks, a higher singles index means more single people. Cost of Living is the opposite - the higher the cost of living, the less desirable the city. To deal with this, I took a scaled reciprocal of the CoL index. This has the effect of translating low CoL indexes into higher scores and vice versa - a city with a really low CoL index of 50 would end up with a score of 200; a city with a really high CoL index of 150 would end up with a score of about 67.