<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Slack Jack on Medium]]></title>
        <description><![CDATA[Stories by Slack Jack on Medium]]></description>
        <link>https://medium.com/@slackjack314?source=rss-f39749f7172------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*htvVGU23tHBTYnDZ__POtw.jpeg</url>
            <title>Stories by Slack Jack on Medium</title>
            <link>https://medium.com/@slackjack314?source=rss-f39749f7172------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 22:42:10 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@slackjack314/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Which high schools send the most students to UCLA?]]></title>
            <link>https://medium.com/@slackjack314/which-high-schools-send-the-most-students-to-ucla-7d6d7820565f?source=rss-f39749f7172------2</link>
            <guid isPermaLink="false">https://medium.com/p/7d6d7820565f</guid>
            <category><![CDATA[college-admissions]]></category>
            <category><![CDATA[college]]></category>
            <category><![CDATA[ucla]]></category>
            <dc:creator><![CDATA[Slack Jack]]></dc:creator>
            <pubDate>Fri, 25 May 2018 17:34:48 GMT</pubDate>
            <atom:updated>2018-05-25T17:34:48.313Z</atom:updated>
            <content:encoded><![CDATA[<p>And for UCLA</p><ol><li>Here are <strong>the top 20 “feeder” schools</strong> to UC Berkeley for the past five years (2012–2016):</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/724/1*K_i3MbChQq9SXInr5WF1Dw.jpeg" /></figure><p>The top 3 enrolled are Diamond Bar (LA Area), Lowell (SF Bay Area), and Torrey Pines (SD Area).</p><p>Compared to Berkeley, there are more schools from Souther California (no surprise). But it’s interesting that the top 3 in terms of admitted are Monta Vista, Mission San Jose, and Lynbrook, all from the Bay Area.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.google.com%2Fmaps%2Fd%2Fembed%3Fmid%3D1C7mgdYusAtnN4UzozTMcqzDWL_s%26hl%3Den_US&amp;url=https%3A%2F%2Fwww.google.com%2Fmaps%2Fd%2Fviewer%3Fmid%3D1C7mgdYusAtnN4UzozTMcqzDWL_s%26hl%3Den_US&amp;image=https%3A%2F%2Fwww.google.com%2Fmaps%2Fd%2Fthumbnail%3Fmid%3D1C7mgdYusAtnN4UzozTMcqzDWL_s%26hl%3Den_US&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=google" width="600" height="450" frameborder="0" scrolling="no"><a href="https://medium.com/media/bed51f2dca85e75bd2d1aad9b5eb6b46/href">https://medium.com/media/bed51f2dca85e75bd2d1aad9b5eb6b46/href</a></iframe><p><strong>2. But who actually gets admitted to UCLA</strong>?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/710/1*b7cbwOCJtzI0_CxjqLJ9Mg.jpeg" /></figure><p>Wow. Very different from Berkeley. The top 20 schools with the highest admission rates are predominately foreign and outside of California. And they’re from all over the world:</p><ul><li>Asia: Victoria Junior College, Raffles Institution, Anglo-Chinese Independent (Singapore); St Pauls Coed Collegiate (Hong Kong), Hankuk, Seoul University, Daewon Foreign Language (South Korea); Dhirubhai Ambani (India)</li><li>Europe: SEven Oaks School (UK); Athens College (Greece)</li><li>Middle East: Dubai College (UAE)</li></ul><p>(the average admission rate for this time period was 17.3%)</p><p>What’s interesting is the low yields for a lot of these schools. So maybe it’s better to take a look at the top foreign schools sending students to UCLA:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/722/1*PtnSvb_On4ca-FowyCnmWQ.jpeg" /></figure><p>Every school on this list is from Asia (including India).</p><p><strong>3. Admitted students from outside California don’t seem to like UCLA as much. </strong>Their yield rates tend to be lower.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/924/1*fwLAhzotAxjpcBH2N5ZXXQ.png" /></figure><p>We see a similar pattern to Berkeley here. For the top 50 California and top 50 Non-California:</p><p>Average admission rate:</p><ul><li>CA: 19.9%</li><li>Non-CA: 28.1%</li></ul><p>Average Yield:</p><ul><li>CA: 30.6%</li><li>Non-CA: 10.0%</li></ul><p><strong>4. Students from private schools also have lower yield rates</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/935/1*jn_bX4A5hHBNIKLsk3rzNw.png" /></figure><p>Average admission rates:</p><ul><li>CA Public: 19.4%</li><li>CA Private: 21.4%</li></ul><p>Average yield:</p><ul><li>CA Public: 33.4%</li><li>CA Private: 28.0%</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7d6d7820565f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Wage Gap for Employees of the City of SF: Ethnicity and Gender]]></title>
            <link>https://medium.com/@slackjack314/wage-gap-for-employees-of-the-city-of-sf-ethnicity-and-gender-1b01e30c0ab0?source=rss-f39749f7172------2</link>
            <guid isPermaLink="false">https://medium.com/p/1b01e30c0ab0</guid>
            <category><![CDATA[gender-equality]]></category>
            <category><![CDATA[wage-gap]]></category>
            <category><![CDATA[san-francisco]]></category>
            <category><![CDATA[race]]></category>
            <dc:creator><![CDATA[Slack Jack]]></dc:creator>
            <pubDate>Mon, 11 Dec 2017 22:20:15 GMT</pubDate>
            <atom:updated>2017-12-13T19:17:30.087Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Curious about how the government of San Francisco is doing when it comes to wage equality? I am too!</em></p><p>It’s usually difficult to get detailed information on wage equality. Most of the <a href="http://fortune.com/2017/06/29/google-2017-diversity-report/">stats</a> I’ve seen published are at the summary level and often don’t include information about pay. The City of SF, however, has made detailed information available. A fantastic dataset on <a href="https://www.kaggle.com/kaggle/sf-salaries">Kaggle</a> contains detailed pay statistics for employees of the city of San Francisco from 2011–2014. For each one of these years, it provides information such as pay stats (e.g., base pay, overtime, etc.), job title, and employee name for the 30,000+ people that the city of SF employees. Yes, it includes employee name.</p><p>Some quick factoids:</p><ul><li>The highest paid person during this time period was Nathaniel Ford, General Manager of SFMTA. He made $567,595 in 2011. This appears to include an almost <a href="http://sanfrancisco.cbslocal.com/2011/06/15/sf-muni-chief-nate-ford-gets-nearly-400000-golden-parachute/">$400k golden parachute</a>.</li><li>SF Supervisors make $109,388 (2014)</li></ul><p>While there isn’t any information on ethnicity, we can make a guess by looking at the employee last name. How? The US Census published information on “<a href="https://www.census.gov/topics/population/genealogy/data/2010_surnames.html">Frequently Occurring Surnames</a>”. For every surname that appears over 100 times in the 2010 Census and likelihood that surname is of a certain ethnicity. Pretty cool.</p><p>For example, 92% of all people who reported the name “Garcia” reported their ethnicity as Hispanic in the 2010 US Census. Similarly, 96% of all “Nguyens” reported themselves as Asian. 95% of “Hansen”s are White.</p><p>(Interestingly, “Lee”, which in the Bay Area is commonly thought of as an Asian last name, has only a 42% likelihood of being Asian in the US. And names such as “Johnson” and “Smith” aren’t that predictive of ethnicity because of the overlap between White and Black people who reported these names.)</p><p>If you match the last names in the SF city data against the US Census data, you should be able to get an estimate of how different ethnic groups are doing in terms of wage equality. I did this, taking only the names that are at least 90% likely to be of a certain race. 90% seems to be a reasonable threshold for using the last name to approximate ethnicity. For example, here are some 90+% names:</p><ul><li>White: Walsh, Sakowski, McDonnell, Adelman, Gowan, Eagan</li><li>Black: Ogbu, Tsegay, Ogbuehi, Nuru</li><li>Asian: Leung, Wu, Tam, Cheung, Kwan, Patel</li><li>Hispanic: Chavez, Gonzalez, Lopez, Castellanos</li></ul><p>(<a href="https://www.facebook.com/notes/facebook-data-science/how-diverse-is-facebook/205925658858/">Facebook</a> used this in approach back in 2009 to estimate the diversity of their user base).</p><p>Here are the results:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/330/1*XnOU5BKD_Fq-v_iK0LRiIw.jpeg" /><figcaption>Mean Total Pay by Ethnicity</figcaption></figure><p>Wow. I didn’t expect that. Asian, Black, and Hispanic employees all make around 20% than white employees (on average). According to this data, Asians make the lowest in total pay — 23% less than Whites. This is curious since it is commonly believed that Asians have high average incomes. Let’s take a closer look at how Asians compare to Whites:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/532/1*SPoEL_b5iRm7LWn3uSZvKQ.png" /><figcaption>SF Employee Pay Distribution: Asian (blue) and White (green)</figcaption></figure><p>Proportionately, White employees are overrepresented at the top end of the pay scale. In fact, 38% of white employees make over $100k per year while only 21% of Asian employees do so. In other words, <strong><em>White employees are almost</em></strong> <strong><em>twice as likely to make $100k/year than Asian employees.</em></strong></p><p>The results are similar with Hispanic employees:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/526/1*QZKALcHd84y3TNeFF04ANA.png" /><figcaption>SF Employee Pay Distribution: Hispanic (orange) and White (blue)</figcaption></figure><p>22% of Hispanic employees make over $100k.</p><p><strong><em>In fact, there appear to be two pay patterns: one for whites and one for minorities.</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/532/1*RT5VcDPHRDH4T63D6yGV1g.png" /><figcaption>SF Pay Distribution: Asian (green), Hispanic (orange), Black (purple), White (blue)</figcaption></figure><p>This result is surprising to me given that San Francisco is one of the most diverse and progressive cities in the US. I haven’t looked at this closely, but my guess is that it’s not because people of difference races get paid differently for the same jobs. For example, all SF Supervisors get paid the same:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/571/1*7U8oKJT3S_8D4luH-CTpgQ.jpeg" /></figure><p>“Equal pay for equal work” could be the case for employees of the SF government, but maybe minorities take on different types of work. Or maybe they’re less likely to get promoted.</p><p><strong>What about gender?</strong></p><p>The US Census data only contains last names, so I used a <a href="https://www.cs.cmu.edu/Groups/AI/areas/nlp/corpora/names/0.html">name corpus</a> of first names that was developed for machine learning.</p><p>I divided the SF employees into ones with typically “male” names, typically “female” names, and “uncategorized.” If a first name appears as both male and female (e.g., Pat, Alex, Jessie), it was left as “uncategorized.”</p><p>The average Total Pay during the period 2011–2014:</p><ul><li>Men: $81,903</li><li>Women: $63,953</li></ul><p>Women who work for the SF government make <strong><em>21% less on average </em></strong>than men. This is very similar to the <a href="https://en.wikipedia.org/wiki/Gender_pay_gap_in_the_United_States">reported range of 18–22%</a>. Here’s what the distribution looks like:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/422/1*ycsoLJiCdCWir1nToxDfFA.png" /><figcaption>Total Pay for Men (orange) and Women (blue)</figcaption></figure><p>Look familiar?</p><p>These are just inferences based on what a name can reveal about a person’s ethnicity and gender. It would be awesome for someone in the <a href="http://sfdhr.org/equal-employment-opportunity">SF Department of Human Resources</a> to publish the actual numbers.</p><p>An aside for nerds: how can we be sure the employees that were categorized don’t have some kind of bias that would skew the results? I’d like to have someone with a deeper stats background look at this, but here are some initial thoughts in ways the data could be skewed:</p><ol><li>An employee identified as a minority is not a minority. E.g., someone who is identified as Hispanic because their last name is “Garcia” in fact isn’t Hispanic. This could be the case, but we should expect to be correct 90% of the time. I repeated the analysis at 95% and the results are similar.</li><li>If employees with names that are highly likely to be minorities are somehow different from employees of that minority in general. This actually may be the case for black employees. The names that the Census identified as being 90% black appear to be African rather than African-American, so the black employees picked up by the Census lookup may in fact be different than black employees overall.</li></ol><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1b01e30c0ab0" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[BMI, speed, and NFL positions]]></title>
            <link>https://medium.com/@slackjack314/bmi-speed-and-nfl-positions-11f90b407551?source=rss-f39749f7172------2</link>
            <guid isPermaLink="false">https://medium.com/p/11f90b407551</guid>
            <category><![CDATA[scatter-plots]]></category>
            <category><![CDATA[sports-science]]></category>
            <category><![CDATA[nfl]]></category>
            <dc:creator><![CDATA[Slack Jack]]></dc:creator>
            <pubDate>Fri, 17 Nov 2017 19:43:07 GMT</pubDate>
            <atom:updated>2017-11-17T19:43:07.209Z</atom:updated>
            <content:encoded><![CDATA[<p>I’m trying to develop an intuitive understanding of k-nearest neighbors classification so I wanted look into a simple example. Classifying NFL football players by position seems like it could be interesting. I’m not a big football fan, but I do know that the players that start a play close to the ball (like the Center and the players around him) seem to be really big and slow while the people on the outsides (like wide receivers) look like they’re lanky and fast.</p><p>I found a pretty cool dataset that lists out <a href="https://en.wikipedia.org/wiki/NFL_Scouting_Combine">NFL Scouting Combine</a> <a href="http://www.nflsavant.com/dump/combine.csv?year=2015">data</a>. According to wiki, the NFL Scouting Combine is a “week-long showcase occurring every February. . .where <a href="https://en.wikipedia.org/wiki/College_football">college football</a> players perform physical and mental tests in front of <a href="https://en.wikipedia.org/wiki/National_Football_League">National Football League</a> coaches, general managers, and scouts.” The dataset includes basic stats like height, weight, and physical tests like 40-yard dash (speed), bench-press (strength), 3-cone (agility), etc.</p><p>Does the data back up the intuition that there are different positions for big, slow people and skinny, fast people? Here’s the initial scatterplot of a few positions (I picked them because I recognized the names of the positions, or so I thought), using BMI and the 40-yard dash as proxies for size (density) and speed:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/404/1*F5lu7StEkvRIQfmJdLAWFA.png" /></figure><p>Looks like the data does back up the intuition. Offensive Tackles (the ones that line up next to the Center, in the gold in the above graph), do tend to be big and slow and Wide Receivers (green) tend to be skinny and fast.</p><p>I ran this data against a few machine learning algorithms with just these two features and expected it to perform pretty well based on the above scatterplot. Here is the accuracy of train vs. test:</p><ul><li>Multinomial Bayes: 0.30</li><li>Logistic Regression: 0.72</li><li>KNN: 0.80</li><li>SVC: 0.81</li><li>Bernoulli Naive Bayes: 0.30</li></ul><p>0.8 is pretty decent (I think) with only 2 features. But that’s only if I include the 5 positions above. What happens if I include all the positions?</p><p>The accuracy is cut in half. KNN is 0.39 and SVC is 0.43. What gives?</p><p>This gives:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/407/1*gXfwSxev1gKShYSM5xz3rw.png" /></figure><p>While it does seem that certain positions “require” certain bmi/speed combinations, a given bmi/speed range also work for multiple positions. For example, theres quite a bit of overlap between wide receivers and cornerbacks:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/404/1*gSjyO6XOKXIWM-smMq80LQ.png" /></figure><p>Do players shift between the two in the early phase of their careers? That would be interesting to know.</p><p>The Running Backs and Wide Receivers comparison is also interesting:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/404/1*-XIYK5K6wyrUoBRRPHsXLA.png" /></figure><p>They’re both fast (WR’s look to be slightly faster), but have different BMI’s. Recruiting guides from NCSA for <a href="http://www.ncsasports.org/recruiting-tools/college-football-recruiting/football-recruiting-guidelines/wide-receiver">WR</a> and <a href="http://www.ncsasports.org/recruiting-tools/college-football-recruiting/football-recruiting-guidelines/running-back-requirements">RB</a>’s bear this out.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/350/1*bOibbSCjgeC-GFeWlwkaog.jpeg" /></figure><p>There seems to be a clear separation between these two groups:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/391/1*0NsYnVKKJN1nztHusrun4A.jpeg" /></figure><p>What separates these two groups? Turns out the dividing line is the Defensive End:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/403/1*lJEgWP0JoUkcaBOzrjW5bA.png" /></figure><p>I didn’t know there was a difference between Defensive End and the other big folks that line up around the ball, but the Defensive End sets up next to the Defensive Tackle:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/272/1*8zaLlnQ0XRUZRtBrldJtxQ.png" /></figure><p>So being just a few feet away from the ball on defense requires a different bmi/speed range. But the Defensive End lines up across from the Offensive Tackle, and the Offensive Tackles have similar bmi/speed to other linemen closer to the ball. I guess this makes sense because in case a running back breaks free, the defensive ends are the ones that need to chase him down. The Offensive Tackles probably don’t do any of that chasing.</p><p>That’s all.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=11f90b407551" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Which high schools send the most students to UC Berkeley?]]></title>
            <link>https://medium.com/@slackjack314/what-high-schools-send-the-most-students-to-uc-berkeley-5331f19a7a60?source=rss-f39749f7172------2</link>
            <guid isPermaLink="false">https://medium.com/p/5331f19a7a60</guid>
            <category><![CDATA[college-admissions]]></category>
            <category><![CDATA[education]]></category>
            <category><![CDATA[uc-berkeley]]></category>
            <category><![CDATA[affirmative-action]]></category>
            <dc:creator><![CDATA[Slack Jack]]></dc:creator>
            <pubDate>Wed, 11 Oct 2017 22:36:21 GMT</pubDate>
            <atom:updated>2017-10-11T23:44:36.469Z</atom:updated>
            <content:encoded><![CDATA[<p>You want to send your kid to UC Berkeley? I’ve been playing around with University California’s <a href="https://www.universityofcalifornia.edu/infocenter/admissions-source-school">database of admissions data</a> which is awesome and provides, among other awesomeness, admission stats by high school.</p><ol><li>Here are <strong>the top 20 “feeder” schools</strong> to UC Berkeley for the past five years (2012–2016):</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/735/1*gqqeuwPlAKh5T2aukjR16w.jpeg" /><figcaption>Admissions Statistics by School for 2012–2016 combined</figcaption></figure><p><a href="http://www.mvhs.fuhsd.org/">Monta Vista HS</a> in Cupertino tops the list with 204 students enrolled (average of about 40 students per year from a graduating class of about 500). <a href="https://www.fremont.k12.ca.us/Domain/2810">Mission San Jose</a> is just behind with 202 (out of a graduating class of around 550). <a href="http://www.lhs.fuhsd.org/">Lynbrook</a> sent 176 to UC Berkeley (about 35/year of a graduating class of about 450).</p><p>All of these schools are in California, and they’re all public:</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.google.com%2Fmaps%2Fd%2Fembed%3Fmid%3D1CBZ5amRP3bltWibGtOENdCYzD7Y%26hl%3Den_US&amp;url=https%3A%2F%2Fwww.google.com%2Fmaps%2Fd%2Fviewer%3Fmid%3D1CBZ5amRP3bltWibGtOENdCYzD7Y%26hl%3Den_US&amp;image=https%3A%2F%2Fwww.google.com%2Fmaps%2Fd%2Fthumbnail%3Fmid%3D1CBZ5amRP3bltWibGtOENdCYzD7Y%26hl%3Den_US&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=google" width="600" height="450" frameborder="0" scrolling="no"><a href="https://medium.com/media/e6ff30970e4372d6c33a3f322c114339/href">https://medium.com/media/e6ff30970e4372d6c33a3f322c114339/href</a></iframe><p>The entire list is here for people who are interested.</p><p><strong>2. But who actually gets admitted to UC Berkeley</strong>?</p><p>Here is a list of high schools with the highest admission rates to Berkeley</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/741/1*XDcHR9AKgrycPfWF19dgOg.jpeg" /><figcaption>Highest admission rates to Berkeley for 2012–2016. Must have at least 20 admitted students during this time period.</figcaption></figure><p>The overall admission rate during this time period was 15.2%, so these schools do considerably better than average the admission draw. Students at the top schools are more than <em>3 times more likely to be admitted</em> than the average student. There are magnet and private schools, and the list is also more geographically diverse:</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.google.com%2Fmaps%2Fd%2Fembed%3Fmid%3D1j7qJnv1xWi_hVLanEsq-__uneYs%26hl%3Den_US&amp;url=https%3A%2F%2Fwww.google.com%2Fmaps%2Fd%2Fviewer%3Fmid%3D1j7qJnv1xWi_hVLanEsq-__uneYs%26hl%3Den_US&amp;image=https%3A%2F%2Fwww.google.com%2Fmaps%2Fd%2Fthumbnail%3Fmid%3D1j7qJnv1xWi_hVLanEsq-__uneYs%26hl%3Den_US&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=google" width="600" height="450" frameborder="0" scrolling="no"><a href="https://medium.com/media/d1381d963e0182175b5e6a3e6eabf172/href">https://medium.com/media/d1381d963e0182175b5e6a3e6eabf172/href</a></iframe><p>Interestingly, the yield rates of these schools tend to be lower than average (35.9%) is average, which leads us to. . .</p><p><strong>3. Admitted students from outside California don’t seem to like Berkeley as much. </strong>Their yield rates tend to be lower.</p><p>Yield rate is defined as (enrolled students)/(admitted students). A high yield means that when students get in, they tend to go. A low yield means that when students get in, they tend to go somewhere else.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/924/1*C7o0mSW3zpj8euUGgfWRSQ.png" /></figure><p>This graph compares the top 50 CA and top 50 non-CA schools that have the most students admitted to Berkeley. Students from non-CA schools are admitted at a higher rate than those from CA schools (29.5% vs 23.7%). However, Non-CA schools have significantly lower yield than CA schools (20.9% vs. 47.3%).</p><p>Put another way, the lower right hand quandrant in the above graph shows schools where the students are admitted at a higher rate than average, but attend at a lower rate. These are the schools where Berkeley loves them, but they don’t love Berkeley and they’re mostly non-CA schools.</p><p><strong>4. Students from private schools also have lower yield rates</strong></p><p>Among California schools, yield rates tend to be lower for private schools. For the top 50 public and top 50 private schools that admit the most students to Berkeley, here’s a scatter plot of admit rate vs. yield:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/796/1*FbyZKbBjBzlCpOIVzn-wnw.jpeg" /></figure><p>Interestingly, admit rates are *slightly* higher for public than private schools (23.3% vs. 22.9%). Yields, however, are dramatically higher for public schools (48.8% vs. 30.8% for private). The schools in the bottom quadrant are essentially private schools where the students are at higher than average rates, but enroll at lower than average rates. Again, the Berkeley loves them, but they don’t love Berkeley. Maybe the students from these private high schools have more options?</p><p>The upper right quadrant where the love goes both ways. These are mainly public high schools.</p><p><strong>5. Asian students have the highest rates of admission, followed by White students.</strong> Hispanic/Latino and African-American students have the lowest rates of admission. UC Berkeley cannot use race-based criteria for admissions per Proposition 209.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/752/1*YA_AUluJfTEaCX-8ZYWfPw.png" /></figure><p>Remember the top 3 feeder schools? They have a lot of Asian students:</p><ul><li>Monta Vista: 83% (<a href="http://www.mvhs.fuhsd.org/schoolprofile">school website</a>)</li><li>Mission San Jose: 91% (<a href="https://www.usnews.com/education/best-high-schools/california/districts/fremont-unified/mission-san-jose-high-2208/student-body">US News</a>)</li><li>Lynbrook: 87% (<a href="http://lynbrook.schoolloop.com/file/1187658036599/1290582956012/6827653731991825069.pdf">school website</a>)</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5331f19a7a60" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Just how rich are the Bay Area’s neighborhoods?]]></title>
            <link>https://medium.com/@slackjack314/just-how-rich-are-the-bay-areas-neighborhoods-d15f741649?source=rss-f39749f7172------2</link>
            <guid isPermaLink="false">https://medium.com/p/d15f741649</guid>
            <category><![CDATA[income-inequality]]></category>
            <category><![CDATA[housing]]></category>
            <category><![CDATA[san-francisco]]></category>
            <dc:creator><![CDATA[Slack Jack]]></dc:creator>
            <pubDate>Sat, 04 Jun 2016 01:32:16 GMT</pubDate>
            <atom:updated>2016-06-04T01:44:42.118Z</atom:updated>
            <content:encoded><![CDATA[<p>A while back, I saw a list of San Francisco <a href="http://blog.sfgate.com/ontheblock/2014/06/06/s-f-s-richest-zip-codes/">richest neighborhoods</a> by zip code. I remember looking at the number one at the time (94105, which includes South Beach, with a reported median income of $142k) and thinking that it was pretty darn high for a median number. But I also thought that it didn’t seem to represent the extremes of wealth I was seeing around the city and reading about in the news.</p><p>Almost every “the most expensive neighborhood”-type article uses this type of average reporting which can be very misleading. Take <a href="http://sf.curbed.com/2015/11/20/9897826/listing-the-top-ten-most-affluent-zip-codes-in-the-bay-area">this one</a>. According to SF Curbed, the residents of ole’ Atherton make a median income of $220k. I’m sure the number is right, but I also I know quite a few folks that make that much money, and they sure aren’t living in Atherton.</p><p>So I decided to take a look at some tax return data to see if I could come up with some numbers to give a better sense of just how much wealth is in Bay Area neighborhoods. I found a <a href="https://www.irs.gov/uac/soi-tax-stats-individual-income-tax-statistics-zip-code-data-soi">dataset</a> on data.gov which provides the data at a more granular level for 2013 (the most recent year available).</p><p>For each zip code in the US, the dataset provides a breakdown based on income band (under $25k | $25k-50k | $50–75k | $75–100k | $100–200K | over $200k).</p><p>Excellent! This gives us a bit more granularity. In particular, it allows us to look at only those households that make over $200k per year and see how much money the really make, er, report to the IRS. The data also let us see the concentration of these $200k+’ers in a neighborhood — is it just that old wealthy-looking dude that lives around the corner or are they everywhere and we just don’t know it?</p><p>Here are the top 10 wealthiest zip codes in the Bay Area:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/586/1*EPuprjOdWPnDRbcxHUbtXQ.jpeg" /></figure><p>Wow. Just think about these numbers for a second. In 94301/Palo Alto, a full third of the households in that neighborhood make over $200k/year. And of this third, the <em>average</em> (mean) amount they pull just in salary is over $1M in a year. Add other sources of income like capital gains and the amount goes to $3M.</p><p>Using this data, Atherton seems much more like <a href="http://www.mercurynews.com/peninsula/ci_18770819">Atherton</a>. A full 44% of Athertonians make over $200k (not surprising given the $220k number above). And of these, their average salary is over $1M and total income is over $2.5M in 2013.</p><p>(At this point, you may ask <em>“Wait — aren’t you just misleading us with more misleading averages?” </em>We’re still talking about averages, and it’s likely that a few <a href="http://www.marketwatch.com/story/facebook-ceo-mark-zuckerberg-to-raze-four-homes-with-a-view-of-his-own-2016-05-24">really wealthy people</a> in these neighborhoods are drawing the average up. I’d love to get my hands on more granular data, but this is what’s readily available to me. Also, the salary component should be subject to fewer extremes on the high end than total income. But yes, these are still averages, but I think they’re more informative averages.</p><p>Here are the 10 richest San Francisco Zip Codes:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/556/1*EVJUo1wWUEPRAnQyuvtMUg.jpeg" /></figure><p>If you walking around Cow Hollow, 1 in 5 residents there will live in a household that makes over $200k/year. That’s not so hard to believe — there are lots of well-dressed, got-a-good-job-looking people there. But the average amount these folks rake in salary is close to half a million, and total income is $869k. No wonder million dollar homes are selling like pancakes!</p><p>(TBH, I’m not sure what to make of FiDi and Downtown being so high. It doesn’t seem like that many people live in these neighborhoods. But that’s what the data say.)</p><p>The concentration numbers are also lower compared to neighborhoods outside of San Francisco, and that’s to be expected because zip codes in SF tend to traverse multiple socio-economic neighborhoods (e.g., 94115 includes both Pacific Heights and Western Addition). Still, areas like 94123/Marina-Cow Hollow/PacHeights and 94105/South Beach have high concentrations of $200k+ers.</p><p>How does this compare to the US overall? $200k+’ers make up 3.7% of tax returns, and the average income of these households is $509k, $300k of which is salary.</p><p>The Bay Area is even richer that I thought.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d15f741649" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Academy Didn’t Apologize for the Asian Jokes at the Oscars]]></title>
            <link>https://medium.com/@slackjack314/the-academy-didn-t-apologize-for-the-asian-jokes-at-the-oscars-ccf475bc325?source=rss-f39749f7172------2</link>
            <guid isPermaLink="false">https://medium.com/p/ccf475bc325</guid>
            <category><![CDATA[diversity]]></category>
            <category><![CDATA[asian-american]]></category>
            <category><![CDATA[oscarssowhite]]></category>
            <dc:creator><![CDATA[Slack Jack]]></dc:creator>
            <pubDate>Wed, 16 Mar 2016 16:41:15 GMT</pubDate>
            <atom:updated>2016-07-11T15:34:30.435Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/800/1*ksRu4HgdkR1-_VoOKsM9PA.jpeg" /><figcaption>Source: Wikimedia Commons</figcaption></figure><p><em>They issued a statement about cultural sensitivity. But don’t call it an apology.</em></p><p>By now, most people are aware of the tasteless jokes Chris Rock and Sacha Cohen made about Asians during this year’s Oscars.</p><p>Earlier this week, George Takei, Ang Lee, and 22 other members of the Academy wrote <a href="http://www.latimes.com/entertainment/movies/moviesnow/la-et-mn-asians-protesting-oscar-jokes-20160315-story.html">an open letter to the Academy</a> (see text) regarding the targeting of Asians.</p><p>Yesterday, the Academy <a href="http://www.hollywoodreporter.com/race/academy-ceo-writes-apologize-members-876005">responded</a> with the following (yes, this is the full public response):</p><blockquote>“The Academy appreciates the concerns stated, and regrets that any aspect of the Oscar telecast was offensive. We are committed to doing our best to ensure that material in future shows be more culturally sensitive.”</blockquote><p>Let’s be clear: this is not an apology.</p><p>Why? Let’s do a little thought experiment here.</p><p>Say you’re walking down the street, and some random guy punches you in the gut. You want to hit back, but the better angels of your nature prevail and you decide not to. Instead, you ask him to apologize. After a back-and-forth, he finally offers “I regret all forms of violence.”</p><p>Is that an apology? No. An apology would be something like “Sorry for punching you in the gut.” A statement of regret about the <strong><em>specific thing</em></strong> that happened.</p><p>The letter Takei and others wrote was about the specific “targeting of Asians.” The letter was about how “the Oscars show was marred by a tone-deaf approach to its portrayal of Asians” through “tasteless and offensive skits” (namely, the Chris Rock and Sasha Cohen bits).</p><p>In response to a complaint about the specific targeting of Asians, the Academy offered a general, non-specific statement about cultural sensitivity.</p><p>Maybe they thought our understanding of English isn’t nuanced enough to pick up on this verbal sleight-of-hand. Maybe they thought they could get away with it.</p><p>Based on its response, the Academy is not sorry. Not in the least.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ccf475bc325" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Product Management vs. Product Marketing. The struggle is real.]]></title>
            <link>https://medium.com/@slackjack314/product-management-vs-product-marketing-the-struggle-is-real-cccf7ba3f1ea?source=rss-f39749f7172------2</link>
            <guid isPermaLink="false">https://medium.com/p/cccf7ba3f1ea</guid>
            <category><![CDATA[product-marketing]]></category>
            <category><![CDATA[product-management]]></category>
            <category><![CDATA[startup]]></category>
            <dc:creator><![CDATA[Slack Jack]]></dc:creator>
            <pubDate>Thu, 17 Sep 2015 05:19:10 GMT</pubDate>
            <atom:updated>2015-09-17T05:54:39.907Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/649/1*FD30aHaMukb4dgbIUTS0wA.jpeg" /></figure><p>Every so often I get into a discussion with someone about the difference between Product Management and Product Marketing. When an organization has both roles, it seems to inevitably cause some degree of angst. What do Product Marketing does? What does Product Management do? Are Product Managers just supposed to implement what product marketers say? Or do Product Marketers make recommendations for the product manager to decide? This post will try to disambiguate the roles as well as offer a view on product marketing/management configurations.</p><p>The simplest thing to address (and which most people I think understand) are the “outbound” activities of Product Marketing. This is stuff like developing marketing collateral, working with the sales team to develop sales pitches, developing a go-to-market strategy, figuring out which customers segments to go after with a given product, etc. These activities are not intended to influence what is being built. Rather,<strong><em> they take the product as a given and are focused on helping ensure adoption of the product.</em></strong> Most people I talk to understand this set of activities and how they differ from Product Management.</p><p>The Product Marketing/Product Management angst really centers around the question of product direction — <strong><em>who decides what to build. </em></strong>Let’s expand on the leftmost area of the above diagram since that’s where the activities related to the “what to build” question take place:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/718/1*sZyCYtQJ9TH0XjB5bPNLEw.jpeg" /></figure><p>The “in-bound” Product Marketing function (as well as other roles such as “Business Product Management,” and “Product Strategy”; at PayPal in the late 2000s, we had the role “Business Unit Manager”), typically focuses on the front end of this funnel. These activities address issues such as target market segment, customer, vertical strategy, business model, etc.</p><p>Let’s take an example of a company that’s building next generation HR software. At this end of the funnel, a Product Marketer might survey all the major HR software vendors (e.g., SAP/Successfactors, Workday) and other related companies (e.g., Slack, TinyHR) and determine that no one has a good solution to both document and track weekly goals. So a Product Marketer might suggest that the company build a weekly tracking system that integrates into people’s 1:1 meetings with their manager. The Product Marketer may further define small-to-medium business as the target market since large businesses already have incumbent HR systems installed which would be too difficult to displace in the short term (I’m making this up, but I don’t think anyone has developed a product for this that beats email.)</p><p>In the early 2000s, the result of this work was a “Marketing Requirements Document”, or “MRD.” This MRD was a medium-length document (usually 5 to 20 pages) that is supposed to do exactly what it’s called — it describes what the market needs (requires) in a product. The problem was that some MRDs were very high level (e.g., culminating in “a weekly tracking system that integrates into employee’s daily workflows) and others would get very specific (e.g., “status reports should be one single field, plain-text format only, and automatically added to the relevant Google Calendar event immediately upon report submission).</p><p>So if you’re Product Manager or Product Marketer reading this, you can easily start seeing the problem. Shouldn’t a Product Manager be doing at least some of this work? If not, what work should she be doing?</p><h4>#1: Product Marketing alongside Product Management</h4><p>In my experience, companies tend to take one of two approaches. The first is to separate the functions:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/742/1*CvWfgO1_JJKerTNX_VE9eg.jpeg" /></figure><p>This is what PayPal did in 2007. We had Product Marketers (they were called “Business Unit Managers”) who were responsible for the more “market-facing”, strategic questions. They were then coupled with Product Managers who were responsible for working with the development teams to specify and build the actual products. This was<em> </em>hugely problematic.</p><p>While Product Marketing was responsible for strategy (essentially), they had no skin in actually building the product. This created a strange dynamic — are Product Managers supposed to build what the Product Marketers wanted? If so, that relegates the Product Manager to a spec monkey. Or are Product Managers supposed to take the Product Marketer’s ideas as simply recommendations? This gives Product Managers ownership, but then the Product Marketers are more like internal consultants. And to make matters worse, PayPal hired individuals with similar profiles (namely MBAs from good schools) into both roles. So you had two sets of similarly oriented people in different roles, fighting for direction of the product. Product Managers thought Product Marketers didn’t add any value; Product Marketers thought Product Managers didn’t know what they were doing. It was a terribly fraught arrangement, one which resulted in continuous tension and distraction. Management ended up doing some strange things as a result, like creating a new role called “Technical Product Manager” (basically a spec monkey).</p><p>Having said all that, this configuration can work, but usually after a certain level of complexity or size is reached, and only with a clear understanding of who owns product direction. More on that later.</p><h4>#2: Product Marketing and Product Management Responsible for different things</h4><p>The second approach is to scope the Product Manager’s role so that includes the more market-facing, strategic activities:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/475/1*Ijn2JkS8kmPVRhE5pVSuXA.jpeg" /></figure><p>Now I’m a Product Manager, so of course I’m going to like this approach :). In my somewhat biased view, I’ve found that this configuration is much more effective at least up to a point (e.g., company size). The Product Management function should be scoped so that it includes the market and customer-facing elements such as figuring out what market segments to go after, what order to go after them, what customer/user types to prioritize, etc. A Product Manager should own her own strategy. She should be out talking to customers, identifying pain points, developing solutions. The “CEO of your product” is a bit overused, but I do think it’s applicable in this situation. A CEO shouldn’t delegate company strategy and Product Manager shouldn’t delegate product strategy.</p><p>The objection I often hear is that this is too much work for one individual. That’s certainly possible. But I think it’s better to support that individual by adding, say, an analyst to help do things like competitive analysis than it is to create an entirely adjacent function to do the same. And if, as a leader, you don’t have confidence in your Product Manager to handle these responsibilities (in addition to working with the development team to build the product), you should ask yourself why.</p><p>In my view, the the starting point when thinking about potential Product Marketer/Product Management configurations should be:</p><ol><li>A Product Management is responsible for soup-to-nuts product development, including product direction (all the “inbound” activities)</li><li>A Product Marketing is responsible for ensuring customer adoption once the product has been decided upon (all the “outbound” activities)</li></ol><h4>When to go from #2 to #1</h4><p>From what I’ve seen, the #2 configuration makes sense for companies between (<em>very</em> simplistically) 20 and a few hundred employees. Below that level, having Product Marketing and Product Management adjacently trying to answer the “what to build” question results in too many cooks in the kitchen. But once the company reaches a certain size or level of complexity, the first configuration starts making a lot more sense (see <a href="http://www.slideshare.net/gerardodada/defining-product-marketing?related=2">this</a> for a more comprehensive treatment on Product Marketing). There are lots of factors that play into when to make this transition (e.g., complexity of the market, complexity of the product, how much your company is willing to customize a product for specific customers, etc.), and one guiding question should be “is there enough work for two groups?” And when it’s time to make the transition, it’s important be clear about how product direction is set and how the various activities influece product direction. For example, if Product Marketing owns the market segmentation and prioritization, exactly how does this work influence product direction? Some options:</p><ul><li>If it’s consultative, is the Product Marketer okay with that?</li><li>If it’s through negotiation, are both the Product Market and Product Manager clear on the nature of that negotiation?</li><li>If it’s directive, is the Product Manager okay with that?</li></ul><p>One issue I’ve seen is that some companies try to make this transition too early — they want Product Marketing in addition to Product Management when there isn’t enough work around for both. A lot of high-growth companies have the philosophy of hiring ahead of the curve when they can. When talent is scarce as it is in the Bay Area, this can be a very wise approach. But just make sure the ownership of product direction is clear so people spend less time bickering and more time building.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=cccf7ba3f1ea" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>