Doctor, validate thyself!

I recently wrote about my long-term fascination with American electoral geography, the way voting patterns are distributed across states, Congressional districts, counties and other areal units.

Pursuing this interest as an undergraduate political science major, I began to explore state-level presidential voting data. During my junior year, I created a large chart that ranked how states had voted in a series of recent presidential elections, from most to least Democratic, concluding with the 1984 presidential election (then the most recent one).

And I noticed that while Ronald Reagan, the incumbent Republican president, had absolutely walloped Democrat Walter Mondale in 1984, winning the popular vote by 18.2 percentage points (58.8-40.6%) and the Electoral College vote 525-13 (Mondale won only his home state of Minnesota [49.7-49.5%] and the District of Columbia [DC]), there were a few states Mondale lost by a much smaller margin than 18.2 percentage points: Massachusetts (-2.8 percentage points), Rhode Island (-3.6), Maryland (-5.5), Iowa (-7.4), Pennsylvania (-7.4), New York (-8.0) and Wisconsin (-9.2).

As usual, all presidential data are from Dave Leip’s indispensable Atlas of U.S. Presidential Elections.

Consider Pennsylvania, the state in which I was born. While the nation was voting for Reagan by 18.2 percentage points, Pennsylvania was voting for Reagan by “only” 7.4 percentage points (53.3-46.0%), a difference of 10.8 percentage points.

That is, Pennsylvania in 1984 was 10.8 percentage points MORE Democratic than the nation as a whole. Had Mondale lost by “only” 10 percentage points, he would (theoretically) have won Pennsylvania 25 electoral votes (EV), as well as those of Iowa (8), Maryland (10), Rhode Island (4) and Massachusetts (13)—an additional 60 EV.

And had Mondale lost by “only” 7.7 percentage points—as Democrat Michael Dukakis would to Republican George H. W. Bush in 1988—he would also have theoretically won the combined 53 EV of New York (36), Wisconsin (11) and West Virginia (6), boosting his total to 126 EV (better, but still 144 EV shy of the 270 needed to win the White House).

Still, that is close to the 112 EV Dukakis won in 1988.[1] As the purple-inked states on this beautiful hand-drawn map[2] show, Dukakis lost seven states (Illinois, Pennsylvania, Maryland, California, Vermont, Missouri, New Mexico) totaling 125 EV by smaller margins (2.1-5.0 percentage points; mean=3.3) than he did nationally. Had Dukakis lost the election by just 2.7 points, he would theoretically have won 237 EV, only 33 shy of the necessary 270.

1988 Presidential map

The conclusion I drew (no pun intended) was that the “relative partisan margin” of a state—how much more or less Democratic it was than the nation as a whole in a given election—was a useful way to think about electoral geography. Of course, other elections in the state (governor, United States Senate, United States House) are of interest as well, as Paul T. David observed in his Party Strength in the United States, 1872-1970; at one point, I even examined the partisan composition of state legislatures.

Good times.

Two decades later, despite having walked away from a doctoral program in political science, I was still interested in these questions, and I began to collect state-level presidential data again.

My primary goal was to get a sense of how EV’s would be distributed between the parties in the next presidential election (either 2008 or 2012) given a series of hypothetical national popular votes (e.g., Democrat wins nationally by 3 percentage points), essentially updating the exercises with 1984 and 1988 presidential election data I summarized earlier. I was particularly interested in whether the Democratic or the Republican presidential nominee would win more EV if the national vote were divided evenly between the two-major parties.

Having gathered these data, I set about constructing a measure of the relative partisanship of a state, intending to combine data from multiple elections to smooth out any idiosyncratic results.

For example, Democratic presidential nominees won Michigan by an average of 7.4 percentage points from 1992 through 2004, making the state an average 4.3 percentage points more Democratic than the nation. Democrat Barack Obama then won the Wolverine State by 16.4 percentage points in 2008 (9.2 percentage points better than he did nationally). In 2012 and 2016, meanwhile, the average margin in Michigan (with Republican Donald Trump winning by 0.2 percentage points in 2016) dropped to just 4.6 percentage points (only 1.6 percentage points more Democratic than the nation). A reasonable explanation (though not a conclusive one) for the Democratic spike in 2008 is the disproportionate impact of the 2007-08 recession on the automobile industry in Michigan, as voters took out their frustrations with term-limited President George W. Bush on 2008 Republican presidential nominee John McCain.

The questions then became

  1. How many years do I use?
  2. How, if at all, do I “weight” these elections?

My initial instinct was to use five years of data, with a weighting scheme of 1-2-3-4-5, meaning the least recent of the relative Democratic margins (D%-R% of total state vote minus D%-R% of total national vote) would be weighted 1/15 while the most recent one would be weighted 5/15, or 1/3.

This became my first “weighted relative Democratic margin” (W-RDM).

However, as I was also interested in assessing changes in relative state-level partisanship over time, using five elections meant that, prior to 2016, I only had four W-RDM values for a state—giving me only three election-to-election changes in W-RDM to examine[3].

I finally settled on three years in what I call my 3W-RDM[4] in order to minimize the fact that presidential and vice-presidential nominees tend to fare better, relative to their overall performance, in their home states. It is rare for one person to be on at least three consecutive presidential tickets (only two, George H W Bush, 1984-1992 and Gore, 1992-2000, of 21 total unique presidential and vice-presidential nominees, 1984-2012).

And that is the measure I have utilized in a series of posts (here, here, here; I do not specifically use 3W-RDM here, but the logic is the same).

As an example, here is how Nevada voted for president in 2004, 2008 and 2012:

             Year                State D% – R%                      National D% – R%              RDM

             2004                           -2.4                                               -2.5                         D+0.1

             2008                           12.5                                               7.3                         D+5.2

             2012                           6.7                                                 3.9                         D+2.8

The weighted average of the RDM values is (0.1 + 2*5.2 + 3*2.8)/6 = D+3.2. This was Nevada’s 3W-RDM prior to the 2016 election, so one would have expected that year’s Democratic nominee to do 3.2 percentage points better in Nevada than nationwide.

The 2016 Democratic presidential nominee, Hillary Clinton, won the national popular vote by 2.1 percentage points. So, my best estimate (based upon Nevada’s recent voting history) was that Clinton would win Nevada by 5.3 percentage points (2.1+3.2). This estimate was too optimistic, however, as she won Nevada by 2.4 percentage points, 2.9 percentage points lower than expected.


Just bear with me while I briefly describe two other highly reputable approaches to calculating the relative partisan margin of a state (or other areal unit).

The Cook Political Report, the “independent, non-partisan newsletter that analyzes elections and campaigns for the US House of Representatives, US Senate, Governors and President as well as American political trends” has been essential reading for any serious student of American politics since its founding in 1984 by Charlie Cook, formerly “a staffer on Capitol Hill, a campaign consultant, a pollster, and a staff member for a political action committee.”

In 1997, Cook began to calculate the Partisan Voting Index (PVI) as a way to measure “how each [state or Congressional] district performs at the presidential level compared to the nation as a whole.”

The Cook PVI is simply the difference (state minus nation) between two averages:

  1. The average Democratic share of the state-level two-party vote in the previous two presidential elections
  2. The average Democratic share of the national two-party vote in the previous two presidential elections.

In 2008, Obama and McCain won 52.9% and 45.6%, of the national popular vote, respectively, splitting 98.5% of the total vote. Looking only at this two-party vote, Obama received 52.9/98.5 = 53.7% and McCain received 45.6/98.5=46.3%, meaning Obama beat McCain nationally by 7.4 percentage points in the two-party vote.

A similar calculation for 2012 (Obama 51.0%, Republican Mitt Romney 47.1%) shows that Obama beat Romney nationally in the two-party vote by 3.9 percentage points.

The average of 7.4 and 3.9 is 5.7.

In Nevada, meanwhile, overall Obama beat McCain 55.1-42.6%, and he beat Romney 52.4-45.7%; in the two-party vote, Obama won by margins of 12.8 (56.4-43.6%) and 6.8 (53.4-46.6%) percentage points.

The average of 12.8 and 6.8 is 9.8.

Subtracting 5.7 from 9.8 gives you 4.1, meaning that the PVI for Nevada going into 2016 was D+4.1, only a little more Democratic (D+3.2) than the 3W-RDM suggested.

The other approach is the “partisan lean” calculated by the data journalism website, a favorite of this blog.

It is even more straightforward than Cook PVI:

(RDM 2nd-most recent presidential + 3*RDM most recent presidential election)/4

Using Nevada again, we have already seen that in 2008 and 2012, Nevada voted 5.2 and 2.8 percentage points more Democratic than the nation; the 538 partisan lean (PL) formula gives you (5.2 +3*2.8)/4 = (5.2+8.4)/4=13.6/4=3.4.

Thus, Nevada’s 538 PL going into 2016 was D+3.4, broadly similar to the Cook PVI of D+4.1 and the 3W-RDM of D+3.2, and the projected Nevada vote based on the 538 PL was D+5.5.


In this post, I assessed the validity of one of my baseball player performance metrics—the Index of Offensive Ability—by comparing it to two other commonly-used statistics, OPS+ and WAR. Here is how I described validity in that post:

Validity is the extent to which an index/measure/score actually measures what it is designed to measure, or “underlying construct”. While now considered a unitary concept, historically, there were three broad approaches to “assessing” validity: content, construct and criterion.

 Content validity is the extent to which an index/measure/score includes the appropriate set of components (not too many, not too few) to capture the underlying construct (say, a state’s partisan “lean”). Construct validity is how strongly your index/measure/score relates to other indices/measures/scores of the same underlying construct, including a priori expectations of what values should be (sometimes called face validity). Criterion validity considers how well outcomes “predicted” by the index/measure/score align with the actual outcomes.

As you have probably guessed by now, I will spend the rest of this post comparing my 3W-RDM to the Cook PVI and the 538 PL.

But first, I offer a mea culpa.

Before my “Democratic blue wall thesis” post in February 2017, I had used the 3W-RDM (which did not even have a name until then) only for my own edification and amusement. That, however, does not excuse me for not even attempting to validate this measure until now. Moreover, I should not have started writing data-driven posts using the 3W-RDM—implicitly asserting its validity without empirical evidence—until I had performed that validation.

I now present that empirical validation evidence.

Content validity: All three measures not only use presidential election voting data, but they also compare state and national margins in some way. This makes sense because presidential elections feature one party nominee advocating (theoretically) the same platform in every state. By comparison, other statewide elections (governor, Senate) feature candidates who share a party label yet may have very different policy stances. While this may be less true now for Senate races, which are becoming more nationalized, there is still a vast difference between Democratic Senators like Joe Manchin of West Virginia and Elizabeth Warren of Massachusetts, and between Republican governors like Charlie Baker of Massachusetts and Sam Brownback of Kansas.

Thus, despite differences in number of elections utilized, weighting and margin calculation, all three measures arguably have high content validity.

Construct validity. A correlation coefficient (“r”) is a number between -1.00 and +1.00 indicating how two variables co-relate to each other in a linear way[5]. If every time one variable increases, the other variable increases, that would be r= +1.00, and if every time one variable increases, the other variable decreases, that would be r=-1.00. R=0.00 means there is no linear association between the two variables.

I calculated the projected presidential election margin (D% total vote – R% total vote) in each state (plus DC) in every presidential election from 1996 through 2016 by adding each state’s partisan lean score before that election to the actual national popular vote margin. In other words, I repeated the example of Nevada (projected 2016 presidential vote: Cook PVI=D+6.2, 538 PL=D+5.5, 3W-RDM=D+5.2) for all 306 state-level presidential election margins.

Here are the average correlations (PVI vs. PL, PVI vs. 3W-RDM, PL vs. 3W-RDM) between the three sets of projected margins in each election year:

1996    +0.995

2000    +0.994

2004    +0.997

2008    +0.998

2012    +0.997

2016    +0.999

Clearly, each partisan lean measure is nearly identically capturing the underlying partisan distribution of states from most to least Democratic, indicating that each measure has very high construct validity.

Criterion validity. Building upon the analysis of construct validity, the simplest way to assess criterion validity is to compare the projected presidential election margin in each state in each year to the actual margins.

Table 1 does this for each state in 2016. A negative difference means the state voted less Democratic than expected, and a positive difference means the state voted more Democratic than expected. States are sorted from most “less Democratic” to most “more Democratic.”

Table 1: Differences Between Projected and Actual State-Level Presidential Vote Margin (Democratic % – Republican %), 2016

State Cook PVI 538 PL 3W-RDM Mean
West Virginia -17.8% -15.8% -20.0% -17.8%
North Dakota -17.6% -16.2% -16.6% -16.8%
Iowa -13.7% -13.5% -13.5% -13.6%
South Dakota -12.7% -11.6% -12.5% -12.3%
Maine -10.2% -10.2% -10.1% -10.2%
Missouri -10.1% -8.8% -10.7% -9.9%
Indiana -10.8% -9.0% -9.0% -9.6%
Michigan -9.8% -8.8% -9.1% -9.3%
Rhode Island -9.1% -9.4% -9.1% -9.2%
Ohio -8.4% -8.8% -8.9% -8.7%
Montana -8.4% -6.8% -7.4% -7.5%
Wisconsin -7.8% -6.8% -7.1% -7.2%
Hawaii -9.0% -8.6% -3.9% -7.1%
Kentucky -6.5% -6.1% -7.9% -6.9%
Vermont -7.2% -6.9% -5.2% -6.4%
Delaware -7.2% -6.3% -5.8% -6.4%
Wyoming -5.0% -5.0% -6.7% -5.6%
Tennessee -4.5% -4.3% -6.6% -5.1%
Pennsylvania -5.1% -4.7% -5.4% -5.1%
Minnesota -4.1% -4.2% -4.5% -4.3%
New Hampshire -3.8% -3.6% -4.0% -3.8%
Nevada -3.8% -3.1% -2.8% -3.3%
Alabama -2.0% -3.1% -3.3% -2.8%
Mississippi -1.8% -3.3% -2.6% -2.5%
Connecticut -2.9% -2.3% -2.4% -2.5%
Arkansas -1.0% -1.6% -5.0% -2.5%
Nebraska -2.8% -2.4% -1.8% -2.3%
New York -1.8% -2.7% -1.8% -2.1%
South Carolina -0.9% -1.6% -1.4% -1.3%
Oklahoma -0.4% -0.8% -2.1% -1.1%
New Mexico -1.2% -0.6% 0.1% -0.5%
Illinois -0.9% 0.6% 0.2% 0.0%
Florida 0.5% 0.1% 0.1% 0.2%
New Jersey 0.7% -0.6% 0.7% 0.3%
Oregon -0.1% 0.4% 0.6% 0.3%
Louisiana 2.1% 0.5% -0.6% 0.7%
North Carolina 0.8% 0.4% 1.2% 0.8%
Idaho 1.2% 0.9% 0.7% 1.0%
Colorado 1.2% 1.3% 1.9% 1.4%
Kansas 1.8% 2.1% 1.4% 1.8%
Washington 2.9% 3.0% 3.3% 3.1%
Maryland 3.7% 3.1% 4.6% 3.8%
Virginia 3.7% 3.5% 4.5% 3.9%
DC 4.3% 5.2% 4.8% 4.8%
Georgia 5.1% 4.7% 5.1% 5.0%
Massachusetts 5.8% 6.0% 4.7% 5.5%
Alaska 7.2% 3.8% 5.6% 5.5%
Arizona 9.0% 8.0% 7.4% 8.1%
Texas 8.5% 8.4% 8.5% 8.5%
California 9.4% 9.3% 10.6% 9.8%
Utah 24.8% 27.6% 24.8% 25.7%
Mean -2.3% -2.1% -2.3% -2.2%

On average, the measures overestimated Clinton’s performance by a relatively low 2.2 percentage points, with no meaningful difference across measures. Five states—West Virginia, North Dakota, Iowa, South Dakota and Maine—were at least 10 percentage points less Democratic than projected using all three measures; Clinton still won Maine, but by “only” 3.0 percentage points. Four states—Utah, California, Texas and Arizona—were at least seven percentage points more Democratic than projected using all three measures; Clinton won only California of this group, though there are signs that Texas and, especially, Arizona are becoming more Democratic. The massive disparity in Utah  results from the presence of unaffiliated presidential candidate Evan McMullin, a Utah native, on the ballot; his 21.3% of the vote cut deeply into Trump’s vote, so the latter “only” won the state by 17.9 percentage points.

 As Table 2 shows, the performance of these measures—using the average of the actual difference in margins—was the worst since 2000, when they also overestimated Democratic performance by an average of 2.2 percentage points. On average, across all six presidential elections, these measures overestimated Democratic performance by just 0.9 percentage points, a solid performance.

Table 2: Average Difference Between Projected and Actual State-Level Presidential Vote Margin (Democratic % – Republican %), 1996-2016

Year Cook PVI 538 PL 3W-RDM Mean
1996 -0.7% -1.0% -0.9% -0.9%
2000 -2.0% -2.2% -2.5% -2.2%
2004 0.1% 0.4% -0.3% 0.1%
2008 0.7% 0.5% 0.4% 0.5%
2012 -0.7% -0.8% -0.6% -0.7%
2016 -2.3% -2.1% -2.3% -2.2%
Mean -0.8% -0.9% -1.0% -0.9%

These values can be deceptive, however. Consider the performance of the 3W-RDM in 2016. It overestimated Clinton’s margin in Montana by 7.4 percentage points, and it underestimated her margin in Arizona by an identical 7.4 percentage points. In both states the difference was 7.4 percentage points, but averaging the two (0.0 percentage points) would suggest that the 3W-RDM was spot on.

In fact, the three measures missed the actual presidential election margin by at least five percentage points in 26 states.

Table 3 resolves this problem by displaying the average absolute value of the difference between the projected and actual presidential election margins.

Table 3: Average of Absolute Value of Differences Between Projected and Actual State-Level Presidential Vote Margin (Democratic % – Republican %), 1996-2016

Year Cook PVI 538 PL 3W-RDM Mean
1996 5.4% 5.1% 5.6% 5.4%
2000 5.5% 5.9% 6.8% 6.1%
2004 3.9% 3.6% 4.2% 3.9%
2008 6.3% 5.7% 6.2% 6.1%
2012 3.3% 3.2% 3.5% 3.3%
2016 5.9% 5.6% 5.9% 5.8%
Mean 5.0% 4.8% 5.4% 5.1%

On average, the projected and actual presidential election margins differed by 5.1 percentage points in either direction. The 3W-RDM, which differed by an average of 5.4 percentage points, fared slightly worse than the Cook PVI and 538 PL. The best years for these measures were two re-election years, 2004 (3.9 percentage points) and 2012 (3.3), and the worst years were the open seat elections of 2000, 2008 (both 6.1) and 2016 (5.8). The overall worst performance was the 3W-RDM in 2000 (6.8), while the overall best performance was the 538 PL in 2012 (3.2).

I performed identical analyses to those summarized in Tables 2 and 3 using two alternate versions of the 3W-RDM, one which used a 1-3-5 weighting scheme and one which weighted all three years equally. The results were nearly identical to those shown here (though the non-weighted 3W-RDM tended to perform worse on the absolute value differences), suggesting that if the 3W-RDM is slightly less “predictive” than the other two measures, it is not due to the weighting scheme but (most likely) to the inclusion of data from a third election year.

Finally, I counted how many—and which—states were “called” incorrectly by each measure in each presidential election.

Table 4: “Mis-called” States, 1996-2016

Year Cook PVI* 538 PL 3W-RDM Average
1996 9






2000 5






2004 4






2008 4






2012 0 1


0 0.3
2016 5






Mean 4.3 4.5 4.8 4.5

        *States in boldface were “predicted” Democratic wins, and states in italics were

         “predicted” Republican wins.

On average, four or five (out of 51) states are “mis-called” in a given presidential election. Again, the 3W-RDM fared slightly worse (4.8) than average (4.5). Of the 83 total misses (out of 918 possibilities), 52 (62.7%) were states that were projected Democratic wins that were actually won by the Republican nominee.

The presidential election of 1996, when Democrat Bill Clinton cruised to an easy reelection, had the most mis-called states, eight or nine; seven states (Arizona, Colorado, Florida, Montana, North Carolina, South Dakota, Texas) were mis-called by all three measures. By contrast, only one state was mis-called in 2012, Florida by the 538 PL: it projected Obama would lose Florida by 0.1 percentage points when he in fact won it by 0.9 percentage points.

Despite these differences, I would argue that all three measures have high criterion validity, as each does a reasonably good job of “projecting” the actual presidential election margin in a given state and year. My 3W-RDM performed only slightly worse than the other two measures, so I will stick with it for now.


One final note about the utility of partisan lean measures.

The Alabama special Senate election between Republican Roy Moore and Democrat Doug Jones to be held on December 12, 2017 is drawing national attention for two reasons. One, a win by Jones would reduce the Republican Senate majority to 51-49. Two, Moore has been dogged by allegations of sexual misconduct with minors (as well as having been removed twice as Alabama’s Chief Justice for defying federal court orders).

The public polls of this election, which once showed a Moore lead of ~11 percentage points, have tightened considerably since the allegations first appeared on November 9, 2017. As of now, depending on how you aggregate and weight these polls, Moore is somewhere between four percentage points ahead and one percentage points ahead; my best estimate is that Moore is ahead 1.7 percentage points.

But consider this. Following the 2016 presidential election, the average partisan lean for Alabama (using all three measures) is D-28.7. As of this writing, the best estimate of how Democrats will fare in the 2018 Congressional elections is that they are ahead by 7.8 percentage points.

Putting these two values together implies that a generic Republican Senate candidate should be leading a generic Democratic Senate candidate by 20.9 percentage points (28.7 minus 7.8): this should not even be a close contest.

However, the polls suggest that Jones is performing somewhere between 16.9 and 21.9 percentage points better than a generic Democrat—that is a stunning difference, and one that may bode very well for Democrats in 2018.

Until next time…

[1] Technically, he only won 111, as one Democratic elector in Washington (state) cast his presidential vote for Lloyd Bentsen, the 1988 Democratic nominee for vice president, and cast his vice presidential vote for Dukakis.

[2] I freely confess to being the artist. This kid-friendly (fine, I had just turned 22) exhortation to vote must have been in the Comics section of the Washington Post (I was living in DC at the time) the Sunday before the 1988 elections.

[3] My data start in 1984, so I would only have 5W-RDM for 1984-2000, 1988-2004, 1992-2008, 1996-2012 and 2000-2016.

[4] I have experimented with adding a weighted linear trend to the 3W-RDM. The logic is that if I want to use the previous three election margins in a state to “forecast” the state margin in the next election, I should account for the fact that, over time, some states are growing relatively more Democratic (e.g., Nevada has become 11.7 percentage points more Democratic relative to the nation since 1984-1992) or less Democratic (e.g., West Virginia, 44.7 percentage points). Adding a weighted average of all previous election-to-election changes in RDM to a 3W-RDM would, theoretically, account for any increased partisanship over the ensuing four years. For the analyses below, however, there was very little difference between the 3W-RDM and the 3W-RDM+weighted linear trend, so I exclude it.

[5] More formally r = covariance(x,y) divided by SD(x) * SD(y).

Separating the art from the artist

The director David Lynch—who I dressed as this past Halloween—gave this response to a question about the meaning of a puzzling moment toward the end of episode 15 of Twin Peaks: The Return.

“What matters is what you believe happened,” he clarified. “That’s the whole thing. There are lots of things in life, and we wonder about them, and we have to come to our own conclusions. You can, for example, read a book that raises a series of questions, and you want to talk to the author, but he died a hundred years ago. That’s why everything is up to you.”

On the surface, this is a straightforward answer, one Lynch has restated in different ways over the years: the meaning of a piece of art is whatever you think it is. Every individual understands a piece of art through her/his own beliefs and experiences.

I am reminded of a therapeutic approach to the interpretation of dreams that particularly resonates with me.

You tell your therapist what you remember of a dream. The therapist then probes a little more, attempting to elicit forgotten details. The conversation then turns to the “meaning” of the dream. Some therapists may pursue the Freudian notion of a dream as the disguised fulfillment of a repressed wish (so what is the wish?). Other therapists may look to the symbolism of characters and objects in the dream (is every character in a dream really a version of the dreamer?) for interpretation.

Then there is what you might call the Socratic approach; this is the approach that resonates with me. The therapist allows the patient to speculate what s/he thinks the dream means. Eventually, the patient will arrive at a meaning that “clicks” with her/him, the interpretation that feels correct. The therapist then accepts this interpretation as the “true” one.

That the “dreams mean whatever you think they mean” approach aligns nicely with Lynch’s musing is not surprising, given how central dreams and dream logic are to his film and television work.

We live inside a dream

However, there is a subtext to Lynch’s musing about artistic meaning that is particularly relevant today.


The November 20, 2017 issue of The Paris Review includes author Claire Dederer’s essay “What Do We Do with the Art of Monstrous Men?”

I highly recommend this elegant and provocative essay.

For simplicity, I will focus on two questions raised by the essay:

  1. To what extent should we divorce the artist from her/his art when assessing its aesthetic quality?
  2. Does successful art require the artist to be “monstrously” selfish?

Dederer describes many “monstrous” artists, nearly all men (she struggles when cataloging the monstrosity of women, despite how odious she finds the impact of Sylvia Plath’s suicide on her children) before singling out Woody Allen as the “ur-monster.”

And here is where I discern a deeper meaning in Lynch’s “dead author” illustration.

Lynch’s notion that one brings one’s own meaning to any piece of art is premised on the idea that the artist may no longer be able to (or may choose not to) reveal her/his intent.

But that implies that something about the artist is relevant to understanding her/his art. Otherwise, one would never have sought out the artist in the first place.

The disturbing implication is that it is all-but-impossible to separate art from artist.

This is Dederer’s conundrum, and it is mine as well.


A few years ago, a group of work colleagues and I were engaging in a “getting to know each other” exercise in which each person writes down a fact nobody else knows about them, and then everyone else has to guess whose fact that is.

I wrote, “All of my favorite authors were falling-down drunks.”

Nobody guessed that was me, which was a mild surprise.

Of course, the statement was an exaggeration, a tongue-in-cheek poke at the mock seriousness of the process.

Still, when I think about many of the authors I love, including Dashiell Hammett, Raymond Chandler, Edgar Allan Poe, John Dickson Carr, Cornell Woolrich, David Goodis[1]

…what first jumps to mind is that every author I just listed is male (not to mention inhabiting the more noir corners of detective fiction). So far as I know, my favorite female authors (Sara Paretsky, Ngaio Marsh and Agatha Christie, among others) do/did not have substance abuse problems.

Gender differences aside, while not all of these authors were alcoholics, they did all battle serious socially-repugnant demons.

Carr, for example, was a virulently racist and misogynistic alcoholic.

He also produced some of the most breathtakingly-inventive and original detective fiction ever written.

Woolrich was an agoraphobic malcontent who was psychologically cruel to his wife during and just after their brief, unconsummated marriage[2].

He also basically single-handedly invented the psychological suspense novel. More films noir (including the seminal Rear Window) have been based on his stories than those of any other author.

And so forth.

It is not just the authors I admire who are loathsome in their way.

I never ceased to be amazed by the music of Miles Davis, who ranks behind only Genesis and “noir troubadour” Stan Ridgway in my musical pantheon. His “Blue in Green” is my favorite song in any genre, and his Kind of Blue is my favorite album.

But this is the same Miles Davis who purportedly beat his wives, abused painkillers and cocaine, was taciturn and full of rage, and supposedly once said, “If somebody told me I only had an hour to live, I’d spend it choking a white man. I’d do it nice and slow.[3]

Moving on, my favorite movie is L.A. Confidential.

Leaving aside the shenanigans of co-star Russell Crowe, there is the problem of Kevin Spacey, an actor I once greatly respected.

Given the slew of allegations leveled at Spacey, the character arc of his “Jack Vincennes” in Confidential is ironic.

But first, let me warn any reader who has not seen the film that there are spoilers ahead. For those who want to skip ahead, I have italicized the relevant paragraphs.

Vincennes is an amoral 1950s Los Angeles police officer whose lucrative sideline is selling “inside” information to Sid Hudgens, publisher of Hush Hush magazine, reaping both financial rewards and high public visibility. Late in the film, he arranges for a young bisexual actor to have a secret (and then-illegal) sexual liaison with the District Attorney, a closeted homosexual. Vincennes and Hudgens would then catch the DA and the young actor in flagrante delicto.

Sitting in the Formosa Club that night, however, Vincennes has a sudden pang of conscience and leaves the bar (symbolically leaving his payoff—a 50-dollar bill—atop his glass of whiskey), intending to stop the male actor from “playing his part.” Unfortunately, he arrives at the motel room too late; the actor has been murdered.

Determined to make amends, he teams up with two other detectives to solve a related set of crimes, including the murder of the young actor. In the course of his “noble” investigation, he questions his superior officer, Captain Dudley Smith, one quiet night in the latter’s kitchen. Realizing that Vincennes is perilously close to learning the full extent of his criminal enterprise, Smith suddenly pulls out a .32 and shoots Vincennes in the chest, killing him.

OK, the spoilers are behind us.


This listing of magnificent art made by morally damaged people demonstrates I am in the same boat as Claire Dederer: I have been struggling for years to separate art from artist.[4]

And that is before discussing the film that serves as Dederer’s Exhibit A: Woody Allen’s Manhattan.

Dederer singles out Manhattan (still one of my favorite films) because of the relationship it depicts between a divorced man of around 40 (Isaac, played by Allen himself) and a 17-year-old high school named Tracy (Mariel Hemingway).

Not only is the relationship inherently creepy (especially in light of recent allegations by Hemingway and the fact that in December 1997, the 62-year-old Allen married the 27-year-old Soon-Yi Previn, the adopted daughter of his long-time romantic partner Mia Farrow[5]), but, as Dederer observes, the blasé reaction to it from other adult characters in the film makes us cringe even more.

As I formulated this post—having just read Dederer’s essay—I thought about why I love Manhattan so much.

My reasons are primarily aesthetic: the opening montage backed by George Gershwin’s Rhapsody in Blue (and Allen’s voiceover narration), Gordon Willis’ stunning black-and-white cinematography, the omnipresence of a vibrant Manhattan itself.

In addition, the story, a complex narrative of intertwined relationships and their aftermath, is highly engaging. The dialogue is fresh and witty—and often very funny. The characters are quirky (far from being a two-dimensional character, I see Tracy as the moral center of the film) but still familiar.

And then there is the way saw the film for the first time.

The movie was released on April 25, 1979. At some point in the next few months, my father took me to see it at the now-defunct City Line Center Theater (now a T.J. Maxx) in the Overbrook neighborhood of Philadelphia. Given that I was 12 years old, it was an odd choice on my father’s part, but I suspect he wanted to see the film and seized the opportunity of his night with me (my parents had been separated two years at this point) to do so.

City Line Theater

I recall little about seeing Manhattan with him, other than being vaguely bored. I mean, it was one thing for old movies and television shows to be in black-and-white (like my beloved Charlie Chan films), but a new movie?

I do not remember when I saw Manhattan again. At one of Yale’s six film societies? While flipping through television channels in the 1990s? Whenever it was, the film clicked with me that second viewing, and I have only become fonder of it since then.

Two observations are relevant here.

One, it is clear to me that the fact that I first saw Manhattan at the behest of my father, who I adored in spite of his many flaws, heavily influenced my later appreciation of the film[6].

Two, this appreciation cemented itself years before Allen’s perfidy became public knowledge.

These two facts help explain (but not condone) why I still…sidestep…my conscience to admire Manhattan as a work of art.


Ultimately, I think the following question best frames any possible resolution of the ethical dilemma of appreciating the art of monstrous artists:

Which did you encounter first, the monstrous reputation of the artist…or the art itself?

I ask this question because my experience is that once I hear that a given artist is monstrous, I have no desire to experience any of her/his art.

Conscience clear. No muss, no fuss.

That includes not-yet-experienced works by an artist I have learned is loathsome. I have not, for example, seen a new Woody Allen since the execrable The Curse of the Jade Scorpion in 2001.

But if I learn about the artist’s monstrous behavior AFTER reacting favorably to a piece of her/his art, I will often find myself still drawn to the art.[7]

Conscience compartmentalized. Definitely some muss, some fuss.

My love of these works is just too firmly embedded in my consciousness to unwind. Thus, I still love the music of Miles Davis. L.A.Confidential remains my favorite movie. Manhattan may have dropped some in my estimation, but it is still in my top 10.

I am reminded of this line from “Seen and Not Seen” on the Talking Heads album Remain in Light:

“This is why first impressions are often correct.”


And here is where I think Lynch’s impressionistic approach to finding meaning in art and the patient-centered approach to dream interpretation—art and dreams mean whatever we think they mean—relate to the question of loving art while loathing the artist.

Art is a deeply personal experience. The “Authority” Dederer so pointedly disdains in her essay can provide guidance, but (s)he cannot experience the art for you or me.

Put simply, each of us is an “Authority” on any given piece of art—and also on whether or not to seek out that art.

For example:

As a child, I found myself hating The Beatles simply because I was supposed to love them. However, once I discovered their music on my own terms, purchasing used vinyl copies of the “Red” and “Blue” albums (which I still own 30+ years later) along with Abbey Road, The Beatles (the “White” Album), Sgt. Peppers’s Lonely Hearts Club Band, Revolver and Rubber Soul…suffice to say I have 124 Beatles tracks (out of 9,504) in my iTunes, second only to Genesis (288). The Beatles also rank sixth in total “plays” behind The Cars, Steely Dan, Miles Davis (there he is again), Stan Ridgway and Genesis.

Each of us is also the Authority on our changing attitudes toward a given piece of art, including what we learn about the artist, knowledge which then becomes one more element we bring to the subjective experience of art.


Dederer speculates about whether artists (particularly writers) somehow NEED to be monstrous to be successful.

(Upon writing that last sentence, the phrase “madness-genius” began to careen around my brain).

As a writer with advanced academic training in epistemology-driven-epidemiology, I would suggest this study to assess this question.

A group of aspiring artists who had not yet produced notable works would be identified. They would be divided into “more monstrous” and “less monstrous,”[8] definitions to be determined. These artists would be followed for, say, 10 years, after which time each artist still it the study would be defined as “more successful” and “less successful,” definitions to be determined The percentages of artists in each category who were “more successful” would be compared, to see whether being “monstrous” made an aspiring artist more or less likely to be “successful,” or even made no difference at all.

This would not settle the question of the link between monstrosity and art by any means, but it would sure be entertaining.


When Dederer talks about the monstrous selfishness of the full-time writer, she focuses on the temporal trade-offs writers must make—time with family and friends versus time spent writing. Writing is an almost-uniquely solitary endeavor, as I first learned writing my doctoral thesis, and as I continue to experience in my new career.

Luckily, my wife and daughters remain strongly supportive of my choice to become a “writer,” so I have not yet felt monstrously selfish.

There is a different kind of authorial “selfishness,” though, that I would argue is both more benign and more beneficial to the author.

When I began this blog, my stated aim was to focus solely on objective, data-driven stories; my personal feelings and life story were irrelevant (outside of this introductory post).

Looking back over my first 48 posts, though, I was surprised to count 17 (35.4%) I would characterize as “personal” (of which three are a hybrid of personal and impersonal). These personal posts, I observed, have also become more frequent.

Even more surprising was how much more “popular” these “personal” posts were. As of this writing, my personal posts averaged 28.4 views (95% confidence interval [CI]=19.9-36.9), while my “impersonal” posts averaged 14.5 views (95% CI=10.8-18.1); the 95% CI around the difference in means (14.0) was 6.3-21.6.[9]

Moreover, the most popular post (77 views, 32 more than this post) is a very personal exploration of my love of film noir.

In other words, while none of my posts have been especially popular (although I am immensely grateful to every single reader), my “personal” posts have been twice as popular as my “impersonal” posts.

I had already absorbed this lesson somewhat as I began to formulate the book I am writing[10]. Initially inspired by my “film noir personal journey” post, it has morphed into a deep dive not only into my personal history, but also the history of my family (legal and genetic) going back three or four generations.

This, then, is the “selfish” part: the discovery that the most popular posts I have written are the ones in which I speak directly about my own life and thoughts, leading me to begin to write what amounts to a “hey, I really like film noir…and here are some really fun stories about my family and me” memoir-research hybrid. One that I think will be very entertaining.

Whether an agent, publisher and/or the book-buying public ever agree remains an open question.


Just bear with me (I had to write that phrase at some point) while I fumble around for a worthwhile conclusion to these thoughts and memories.

I am very hesitant ever to argue that means justify the ends, meaning that my first instinct is to say that art produced by monstrous artists should be avoided.

But I cannot say that because, having formed highly favorable “first (and later) impressions” of various works of art produced by “monstrous” artists, I continue to love those works of art. I may see them differently, but the art itself has not changed. “Blue in Green” is still “Blue in Green,” regardless of what I learn about Miles Davis, and it is still my favorite song.

And that may be the key. Our store of information about a piece of art may change, but the art itself does not change. It is fixed, unchanging.

Of course, if Lynch and the patient-centered therapists are correct that we each need to interpret/appreciate (or not) works of art as individuals, then how we react to that piece of art WILL change as our store of information changes.

Shoot. I thought I had something there.

Well, then, what about the “slippery slope” argument?

Once we start down the path of singling out certain artists (and, by extension, their works of art) for opprobrium, where does that path lead?

The French Revolution devolved into an anarchic cycle of guillotining because (at least as I understand it) competing groups of revolutionaries began to point the finger at each other, condemning rival groups to death as power shifted between the groups.

This is admittedly an extreme example, but my point is that we once start condemning monstrosity in our public figures, it is difficult to stop.

It is also the case that very few of us are pure enough to condemn others. We all have our Henry Jekyll, and we all have our Edward Hyde, within us. I think the vast majority of us contain far more of the noble Dr. Jekyll than of the odious Mr. Hyde, but we all enough of the latter to be wary of hypocrisy.

And if THAT is not a good argument, then I have one more.

Simply put, let us all put on our Lynchian-therapeutic cloaks and make our own decisions about works of art, bringing to bear everything we know and feel and think, including our conscience…while also understanding that blatant censorship (through public boycott or private influence) is equally problematic…

These decisions may be ethically uncomfortable, but as “Authorities,” they are ultimately ours and ours alone.

Until next time…

[1] Fun fact about Goodis: Philadelphia-born-and-raised, he is buried in the same cemetery as my father.

[2] Woolrich was also a self-loathing homosexual.

[3] This quote is found on page 61 of the March 25, 1985 issue of Jet, in a blurb titled “Miles Davis Can’t Shake Boyhood Racial Abuse.” The quote is apparently from a recent interview with Miles White of USA Today, but I cannot find the actual USA Today article.

As a counter, and for some context, here is a long excerpt from Davis’ September 1962 Playboy interview.

Playboy: You feel that the complaints about you are because of your race?

Davis: I know damn well a lot of it is race. White people have certain things they expect from Negro musicians — just like they’ve got labels for the whole Negro race. It goes clear back to the slavery days. That was when Uncle Tomming got started because white people demanded it. Every little black child grew up seeing that getting along with white people meant grinning and acting clowns. It helped white people to feel easy about what they had done, and were doing, to Negroes, and that’s carried right on over to now. You bring it down to musicians, they want you to not only play your instrument, but to entertain them, too, with grinning and dancing.

Playboy: Generally speaking, what are your feelings with regard to race?

Davis: I hate to talk about what I think of the mess because my friends are all colors. When I say that some of my best friends are white, I sure ain’t lying. The only white people I don’t like are the prejudiced white people. Those the shoe don’t fit, well, they don’t wear it. I don’t like the white people that show me they can’t understand that not just the Negroes, but the Chinese and Puerto Ricans and any other races that ain’t white, should be given dignity and respect like everybody else.

But let me straighten you — I ain’t saying I think all Negroes are the salt of the earth. It’s plenty of Negroes I can’t stand, too. Especially those that act like they think white people want them to. They bug me worse than Uncle Toms.

But prejudiced white people can’t see any of the other races as just individual people. If a white man robs a bank, it’s just a man robbed a bank. But if a Negro or a Puerto Rican does it, it’s them awful Negroes or Puerto Ricans. Hardly anybody not white hasn’t suffered from some of white people’s labels. It used to be said that all Negroes were shiftless and happy-go-lucky and lazy. But that’s been proved a lie so much that now the label is that what Negroes want integration for is so they can sleep in the bed with white people. It’s another damn lie. All Negroes want is to be free to do in this country just like anybody else. Prejudiced white people ask one another, “Would you want your sister to marry a Negro?” It’s a jive question to ask in the first place — as if white women stand around helpless if some Negro wants to drag one off to a preacher. It makes me sick to hear that. A Negro just might not want your sister. The Negro is always to blame if some white woman decides she wants him. But it’s all right that ever since slavery, white men been having Negro women. Every Negro you see that ain’t black, that’s what’s happened somewhere in his background. The slaves they brought here were all black.

What makes me mad about these labels for Negroes is that very few white people really know what Negroes really feel like. A lot of white people have never even been in the company of an intelligent Negro. But you can hardly meet a white person, especially a white man, that don’t think he’s qualified to tell you all about Negroes.

You know the story the minute you meet some white cat and he comes off with a big show that he’s with you. It’s 10,000 things you can talk about, but the only thing he can think of is some other Negro he’s such close friends with. Intelligent Negroes are sick of hearing this. I don’t know how many times different whites have started talking, telling me they was raised up with a Negro boy. But I ain’t found one yet that knows whatever happened to that boy after they grew up.

Playboy: Did you grow up with any white boys?

Davis: I didn’t grow up with any, not as friends, to speak of. But I went to school with some. In high school, I was the best in the music class on the trumpet. I knew it and all the rest knew it — but all the contest first prizes went to the boys with blue eyes. It made me so mad I made up my mind to outdo anybody white on my horn. If I hadn’t met that prejudice, I probably wouldn’t have had as much drive in my work. I have thought about that a lot. I have thought that prejudice and curiosity have been responsible for what I have done in music.

[4] This has actually impacted me directly. Privacy concerns prevent me from using names, but I have had long and painful discussions with people close to me who were either related to, or knew very well, artists whose work they admired but who were/are loathsome human beings.

[5] Purportedly, Allen and his quasi-step-daughter (Allen and Farrow never married) had been having a long-term affair.

[6] And, perhaps, of black-and-white cinematography more generally.

[7] There are exceptions to this, of course. As much as I love the Father Brown stories by G.K. Chesterton, his blatant anti-Semitism has likely permanently soured me on his writing.

[8] Acknowledging that “monstrosity” is not binary, but a continuum. We have all had monstrous moments, and even the most monstrous people have had a moment or two of being above reproach.

[9] Using a somewhat stricter definition of “personal” made the difference even starker.

[10] Tentative title: Interrogating Memory: How a Love of Film Noir Led Me to Investigate My Own Identity.

Final thoughts from what is almost certainly my final APHA meeting

I debuted this blog 11 months ago yesterday as a place to tell what I hoped would be entertaining and informative data-driven stories. Given my proclivity for, and advanced academic training in, quantitative data analysis, the vast majority of my 47 prior posts have involved the rigorous and systematic manipulation of numbers.

But not all data are quantitative. Sometimes they are “qualitative,” or simply impressionistic.

A few weeks ago, I wrote a post about my impending trip to Atlanta to attend the American Public Health Association (APHA) Annual Meeting and Expo. This post served two purposes:

  1. To allow me to archive online:
    1. The full text (minus Acknowledgments and CV) of my doctoral thesis (Epidemiology, Boston University School of Public Health, May 2015)
    2. The PowerPoint presentation I delivered in defense of that thesis (minus some Acknowledgment slides) in December 2014
    3. Both oral presentations I delivered at the APHA Meeting
  1. To explore the idea that the decision to change careers (which I detail here) actually began two years earlier than I thought, with the completion of this doctorate.

I submitted three abstracts to APHA (one for each dissertation study) when I was still looking for ways to jumpstart my health-data-analyst job search (and my flagging interest in the endeavor). I was shocked that any of my abstracts were accepted for oral presentation (if only because I had no institutional affiliation) and quite humbled that two were accepted.

Once they were accepted, though, I felt an obligation to prepare and deliver the two oral presentations, despite the fact that I had decided to embark on a different career path.

(I did, however, truncate the length of my attendance from all four days to only the final two days, the days on which I was scheduled to give my presentations.)

I also recalled how much I used to enjoy attending APHA Meetings with my work colleagues. My first APHA Meeting—Atlanta, October 2001—was also the place I delivered an oral presentation to a large scientific conference for the first time.

APHA 2001


There are two interesting coincidences related to this presentation.

One, I gave this presentation at the Atlanta Marriott Marquis, the same hotel in which I just stayed for the 2017 APHA Meeting[1].

Two, the presentation itself—GIS Mapping: A Unique Approach to Surveillance of Teen Pregnancy Prevention Efforts (coauthored with my then-supervisor)—drew upon a long-term interest of mine: what you might call “geographical determinism,” which is a pretentious way of saying that “place matters.”

To explain, just bear with me while I stroll down a slightly bumpy memory lane.

I have always loved maps—street maps, maps of historical events, atlases, you name it. As a political science major at Yale, I discovered “electoral geography.” At one point while I was working as a research assistant for Professor David Mayhew, I mentioned the field to him.

Hmm, he responded. I should teach a course about that next semester.

He did.

I still have the syllabus.

As a doctoral student at Harvard (the doctorate I did NOT finish), I formulated a theory for my dissertation about why some areas tended to vote reliably Democratic while others tended to vote reliably Republican that was based on the way demographic traits (e.g., race, socioceconomic status [SES], religion) were distributed among an area’s population. The idea was that because everyone has a race AND an age AND a gender AND a SES level AND a religion AND so on, the areal distribution of these traits makes some more politically salient than others in that area.

Well…it all made perfect sense to me back in the early 1990s.

Because this was not already complicated enough to model and measure, I originally chose to test this theory using data from presidential primary elections, with all of their attendant flukiness. I even spent a pleasant afternoon in Concord, New Hampshire collecting (hand-written) town-level data on their 1976 presidential primary elections.

Did I mention that New Hampshire has 10 counties, 13 cities, 221 towns, and 25 unincorporated places?

From the start, however, it was an uphill battle getting this work taken seriously[2]. One of the four components of my oral exams in May 1991 was a grilling on the electoral geography literature review I had recently completed.

Rather than ask me questions about (for example) J. Clark Archer’s work on the geography of presidential elections, however, the professor who would soon chair my doctoral committee peppered me with questions about why we should study political/electoral geography when academic geography departments were closing or what James Madison’ antipathy to faction said about viewing elections through the lens of geography.

I have no recollection of how I answered those questions, but I know that I passed those exams by the skin of my teeth[3].

Ironically, just nine

The real kicker, though, came a year later.

Harvard at the time had a program with a name like “sophomore seminars.” These small-group classes were a chance for doctoral students to prepare and teach a semester-length seminar of their own design to undergraduate political science majors.

I eagerly jumped at the chance and applied to teach one in American electoral geography, drafting a syllabus in the process. Once it was accepted, I organized the first class, including getting permission to copy a Scientific American article, which I then made copied.

Towards the end of the summer, they posted (I do not remember where, but it was 1992, so it was literally a piece of paper tacked to a bulletin board) the names of the students who would be taking each seminar.

I looked for my class.

I could not find it.

I soon discovered why. Only one student had signed up (and it was not even her/his first choice), so the seminar had been cancelled.

That was one of the most crushingly disappointing moments of my life.

In retrospect, this was most likely when my interest in completing this doctoral program began to seriously wane—even though I stuck it out for three more years.

(In a bittersweet bit of irony, five years after I walked away from that doctoral program came the 2000 U.S. presidential election. Because of the month-long Florida recount, the “red state-blue state” map of the election burned into the public consciousness. Electoral geography, at least at this very basic level, suddenly became a “thing.” To this day, there is talk of “red,” “blue” and even “purple” states.)

The good news was that the idea of looking at data geographically still appealed to me tremendously, and I was lucky enough to be able to learn and use ArcGIS mapping software in my first professional job as a health-related data analyst. The best moment in this regard there came when I produced a town-level map of alcohol and substance use problems in Massachusetts. The towns with the most severe issues were colored in red, and I noticed that they followed two parallel east-west lines emanating from Boston, and that they were crossed by a north-south line in the western part of the state.

Oh, I exclaimed. The northern east-west line is Route 2, the southern east-west line is I-90 (the Massachusetts Turnpike) and the intersecting north-south line is I-91. Of course, these are state-wide drug distribution routes.

Three professional positions later, temporarily living in Philadelphia, I was doing similar work, but now in the area of teen pregnancy–which brings us back to the oral presentation I delivered late on the afternoon of November 7, 2017 and to the second coincidence.

Its title was “Challenges in measuring neighborhood walkability: A comparison of disparate approaches,” and it was the second presentation (of six) in a 90-minute-long session titled Geo-Spatial Epidemiology in Public Health Research.

In other words, 16 years after my first APHA oral presentation, in the same city, I was once again talking about ways to organize and analyze data geographically.

And while the five-speaker session in which I spoke the following morning (Social Determinants in Health and Disease) was not “geo-spatial,” per sé, the study I discussed (“Neighborhood walkability and depressive symptoms in black women: A prospective cohort study”) did feature a geographic exposure.


I again coauthored and delivered oral presentations at the APHA Meetings in 2002[4] (Philadelphia) and 2003 (San Francisco); for the 2004 Meeting (Washington, DC) I prepared a poster which I displayed along with a woman I supervised.

That talented young woman—now one of my closest friends—was a huge reason why the 2003 APHA Meeting in San Francisco was so memorable. Other, of course, than the fact that it was IN SAN FRANCISCO!





As much as fun as it was to wander through the exhibit halls and chat with the folks from schools of public health, research organizations, public health advocacy groups, medical device firms and so forth; to amass a full bag of free goodies (“swag,” I prefer to call it) in the process; to read and ask questions about scientific posters; and to sit in a wide range of scientific sessions…

(no, I am serious. I really used to enjoy that stuff, especially in the company (during the day and/or over dinner and drinks in the evenings) of friendly work colleagues)

…after about two days, my colleague and I had had enough.

So we literally played hooky from the Meeting one day.

First, I dragged the poor woman on a “Dashiell Hammett” tour, which took place only a few blocks from our Union Square hotel.



Then, we meandered through Chinatown (whose entrance was mere steps away)—stopping for bubble teas along the way—all the way to Fishermen’s Wharf.


Our ultimate destination was the ferry to Alcatraz. The Alcatraz tour may have been the highlight of that trip. That place is eerie, creepy and endlessly fascinating.


Someday I will take my wife and daughters there.

That Meeting was also the apex of my APHA experiences. After three years of them, the 2004 version in DC felt stale. I skipped the 2005 APHA Meeting in Philadelphia, as I had just returned to Boston to start my master’s program in biostatistics at Boston University, though I did briefly attend the 2006 APHA Meeting since it was in Boston, and it was a chance to see former work colleagues.


Ultimately, then, attending the 2017 APHA Meeting in Atlanta was a life experiment, a way to gather qualitative “data” to assess the notion that I had put a health-related data analysis career behind for good.

I arrived in Atlanta on the evening of November 6 and took a taxi to the Marriott Marquis.

Holy moley, is this place huge…and it had those internal glass elevators which allow passengers to watch the lobby recede or approach at great speed.


It was both liberating and lonely not to have work colleagues attending with me. As great as it was not to have to report to anybody, it also meant my time was far more unstructured (other than attending the sessions in which I was presenting).

On Tuesday morning, I dressed in my “presentation” clothes and made my way to the Georgia World Congress Center. This meant taking a mile-long walk in drenching humidity carrying a fully-packed satchel because the APHA chose to reduce its carbon footprint by eliminating shuttle buses.

So I was a sweaty mess when I arrived at the heart of the action. Still, I soldiered on, registering and then checking the location of my session room (luckily, both of the my sessions were in the same room—if only because it allowed me, on Wednesday morning, to retrieve the reading glasses I had left on the podium Tuesday evening).

This place was also massive and labyrinthine. It took me a good 30 minutes just to locate the Exhibit Halls.

I wandered through them for an hour or so, talking to some interesting folks and reading a couple of posters. The swag was wholly uninspiring, I am sorry to say.

And I felt…nothing.

No pangs of regret.

No overwhelming desire to return to this field of work.

No longing for work colleagues (other than a general loneliness).

In fact, I mostly felt like a ghost, the way one sometimes does walking around an old alma mater or place you used to live.

This was my past, and I was perfectly fine with that[5].

That is not to say I did not enjoy giving my talks (which were very well received—I am usually nervous before giving oral presentations…until I open my mouth, and the performer in me takes charge). I did, very much. I also enjoyed listening to the nine other speakers with whom I shared a dais. I picked up terms like “geographic-weighted regression” I plan to explore further. I even took the opportunity to distribute dozens of my new business cards (the ones that describe me, tongue somewhat in cheek, as “Writer * Blogger * Film Noir Researcher * Data Analyst”).

But none of that altered my conviction that I have made the right career path decision. I have no idea where the writing path will ultimately lead (although the research for my book has already taken me down some unexpected and vaguely disturbing alleys), professionally or financially, but I remain glad I chose that path.

One final thing…or perspective.

Tuesday, November 7 was also the day that governor’s races were held in New Jersey and Virginia, along with a mayor’s race in New York City and a wide range of state and local elections nationwide.

I had expected to settle in for a long night of room service and MSNBC viewing, but the key races were called so early that I decided to take quick advantage of the hotel swimming pool.

Yes, I waited at least 30 minutes after eating to enter the water.

The pool at the Atlanta Marquis Marriott is primarily indoors (and includes a VERT hot hot tub, almost—but not quite—too hot for me), but a small segment of it is outside; you can swim between the two pool segments through a narrow opening.

If you look directly up from the three shallow steps descending into the outdoor segment of the pool, you see this (if you can find the 27th floor, one of those windows was my room):


I literally carried my iPhone into the pool to take this photograph, leaning as far back as I could. Thankfully, I did not drop my iPhone in the pool.

Until next time…

[1] The coincidence is not perfect, though, as I do not think we STAYED at the Marriott Marquis in 2001.

[2] Other than the fact that I was awarded a Mellon Dissertation Completion Fellowship in 1994. It was kind of a last-ditch spur to completion. It did not work.

[3] This was the same professor who proclaimed as an aside in a graduate American politics seminar that if you really want to do something hard, get a PhD in epidemiology. Which, of course, I did…25 years later.

[4] Where the Keynote Address was delivered—passionately and to great applause—by an obscure Democratic governor of Vermont named Howard Dean, whose presidential campaign I supported from that moment.

[5] The one caveat to this blanket page-turning is my ongoing interest in the geographic determinism, which I am indulging through state- and county-level analyses of the 2016 presidential elections. This may be the one successful way to lure me back into the professional data-analytic world.

The 2016 U.S. presidential election viewed through one statistic

The 2016 United States (U.S.) presidential election is one of those elections (1948, 1960, 1968 and 2000 also come to mind) people will be re-hashing as long as the U.S. continues to HAVE presidential elections. I have already shared data-driven thoughts on the 2016 U.S. presidential election here, here, here, here, here and here.

Grounding my thoughts about this election is the following sequence of data points (drawn from Dave Leip’s invaluable Atlas of U.S. Presidential Elections):

  • Democratic presidential nominee Hillary Clinton won 2,868,518 more votes OVERALL than Republican presidential nominee Donald Trump (48.0% vs. 45.9%).
  • Trump won the election because he won more Electoral College votes (EV; 306 to 232[1])
  • Trump won more EV because he won narrow victories in three states:
    • Pennsylvania (20 EV): 44,292 votes, or 0.72%
    • Wisconsin (10 EV): 22,748, or 0.76%
    • Michigan (16 EV): 10,794, or 0.22%
  • Trump won because of just 77,744 votes in three closely-fought states, or 0.057% of the 137,125,484 votes cast in the 2016 U.S. presidential election.

I want to shout these numbers whenever political pundits or elected officials and their allies fret about “how Democrats can ever win back voters in 2018 or 2020.”

To all those folks I say, Chill! The 2016 U.S. presidential election was VERY close, not to mention that Democrats also netted two U.S. Senate seats and six U.S. House of Representatives seats that year.

And while it is absolutely true that, relative to the extraordinarily Democratic years of 2006 and 2008, Democrats have been losing ground badly at the state level (with 2017 election results suggesting a slow-moving reversal), that is not the focus of this post.

Instead, I want to focus on the single statistic that strikes me as the key to understanding the outcome of the 2016 U.S. presidential election.


First, however, just bear with me while I briefly address “electoral legitimacy” arguments made about that election.

These basically fall into two groups:

  1. Russian cyberattacks amplified through American social and traditional media
  2. Voter suppression efforts

The goal of the Russian cyberattacks (including, but not limited to, hacking Democratic National Committee e-mails and releasing them through WikiLeaks; purchasing thousands of ads on social media platforms; coordinating “trolling” on those same social media platforms by Russian nationals) appears to have been to sow discord in the American electorate; punish 2016 Democratic presidential nominee Hillary Clinton; and, PERHAPS, promote the candidacy of Republican presidential nominee Donald Trump (with or without “collusion” on their part).

That such meddling did occur is widely accepted, even if the efficacy of that meddling is debatable.

But the next question to be asked is this: as a result of this interference, how many voters who would otherwise have voted for Clinton did not vote for her, regardless of whether they voted for somebody else or simply did not cast a presidential vote at all?

This counterfactual may not be possible to assess given the voting data at our disposal and the multitude of reasons we choose one candidate over another.

Well, besides simple partisanship that is (data source found by clicking on election year):

Table 1: Percentage of Self-Identified Partisans Who Voted For Presidential Candidate of Their Party, 2000-2016

Election % Democrats

voting Democratic

% Republicans voting Republican Margin among Independents
2016 89%

(36% of electorate)



+4% Republican


2012 92%




+5% Republican


2008 89%




+8% Democratic


2004 89%




+1% Democratic


2000 87%




+2% Republican


Mean 89%






In the previous five presidential elections, 87-92% of self-identified Democrats voted for the Democratic nominee, and 88-93% of self-identified Republicans voted for the Republican nominee. Self-identified Independents (whose share of the electorate seems to be increasing over time), most of whom usually cast their ballots for the same party over time, divided their votes fairly evenly between the Democratic and Republican nominees (while also being more likely to choose a third-party option[2]) over these same elections.

American politics is highly polarized, and the vast majority of voters simply vote for the nominee with the same party identification as them, so the pool of voters who would have been swayed by Russian interference was already very small.

Again, that is not to say the meddling did not occur, that it was not an attack on our sovereign democracy, and that no votes were changed from “Clinton” to either “not Clinton” or a non-vote. I just think there is a far less “conspiratorial” way to understand the results of the 2016 U.S. presidential election.

As for voter suppression efforts like restrictive voter ID laws, fewer polling places and shorter/no early voting periods, there is some evidence that this occurred in states highly relevant to the outcome of the 2016 U.S. presidential election, including Wisconsin and North Carolina.

Yes, I wrote “North Carolina.”

While the “path of least resistance” for Clinton would have been to flip just under 78,000 votes in three “Rust Belt” states, an alternate path would have been to flip just 285,826 votes (0.21%) in two southeastern states: Florida (Clinton -112,911, or 1.2%) and North Carolina (Clinton –172,915, or 3.6%). Or to flip 157,203 votes (0.11%) in Florida and Pennsylvania…you get the idea.

But, even IF Wisconsin and North Carolina had voted for Clinton if voter suppression had not existed (a difficult counterfactual to prove), that would only have garnered Clinton 25 additional EV, increasing her total to 257, 13 shy of the 270 required for victory. She would still have needed to win one of Michigan, Pennsylvania or Florida, states where there have been no claims of voter suppression of which I am aware.

The point is, while Russian interference and voter suppression certainly happened, demonstrating that they prevented enough votes for Clinton in the right combination of states to deny her an Electoral College victory in the 2016 U.S. presidential election is extremely difficult. The simple fact that each was attempted is pernicious enough.


What makes the 2016 U.S. presidential election stand out from the pack is how disliked both major party nominees were.

totalfavunfavehorizontalAccording to the exit polls, Clinton was viewed favorably by 43%, and unfavorably by 55%, of the 2016 presidential electorate; the corresponding values for Trump were 38% and 60%, respectively. These line up nicely with the RealClearPolitics (RCP) averages going into Election Day (November 8, 2016): Clinton 42%/54%, Trump 38%/58%.[3].

On average, 95% of those with a favorable view of a candidate voted FOR that candidate. Among voters with an unfavorable view of Clinton, 81% voted for Trump, and among voters with an unfavorable view of Trump, 77% voted for Clinton.

Here is the kicker, however:

An unusually high 18%[4] of the electorate had an unfavorable view of both Clinton AND Trump. This pivotal portion of the electorate gave 47% of their votes to Trump, 30% to Clinton and 23% to neither candidate.

That’s right, Trump won by 17% percentage points nationwide among voters who disliked BOTH major-party candidates.

And the support for Trump among this portion of the electorate was much stronger in the six states Clinton lost by less than four percentage points (total EV=99):

Table 2: Favorability Ratings for Clinton and Trump in Six Key States, 2016

State EV Trump Margin Clinton Trump Both Unfavorable Margin among

Both Unfavorable

MI 16 +0.2% 42/56 39/59 20% Trump +21%
PA 20 +0.7% 42/57 42/56 17% Trump +25%
WI 10 +0.8% 42/56 35/64 22% Trump +37%
FL 27 +1.2% 45/53 41/57 14% Trump +37%
AZ 11 +3.5% 41/57 41/57 18% Trump +17%
NC 15 +3.6% 43/56 41/58 16% Trump +36%
Mean 16 +1.7% 43/56 40/59 18% Trump +29%

On average, 18% of the voters in these six states had an unfavorable view of both Clinton and Trump, with Clinton earning 27% of their votes (3 percentage points lower than nationwide) and Trump earning 56% of their votes (9 percentage points higher than nationwide). Third-party candidates did worse (18%), on average, than nationwide (23%) with this group in these six states; the exception is Arizona (29%), neighbor to the west of 2016 Libertarian presidential nominee Gary Johnson’s home state of New Mexico[5].

In fact, Trump received an astonishing 60% of the “pox on both your houses” votes in Wisconsin, 61% in Florida and 62% in North Carolina.

I can find no historical data to which to compare these numbers, so I do not know what a typical vote distribution among this segment of the electorate is. Still, it is important to keep in mind that the 2016 U.S. presidential election took place after eight years with one party (Democrats) occupying the White House and no incumbent running. Voters often look to change White House control in these elections: prior to 2016, of the six such elections starting with 1960, the party not occupying the White House had won five of them (1960, 1968, 1976, 2000, 2008). The exception was 1988, when Republican nominee George H. W. Bush beat Democratic nominee Michael Dukakis by 6.8 percentage points and 315 EV.

These elections also tend to be very close, with the party not occupying the White House winning the two-party vote by an average of just 0.3 percentage points and 22 EV (excluding 1988, these values are 1.9 percentage points and 90 EV)[6].

According to the RCP average, voters on Election Day 2016 felt the country was going in the wrong direction by a margin of 61-31%. Combine this with an eight-year/no-incumbent election and Clinton (or any Democratic presidential nominee) should always have been seen as a slight underdog. The historic unpopularity of Trump (net -21 percentage points) may have led observers to conclude that this election would be different, but they did not take into account Clinton’s only-marginally-better favorability rating (-13 percentage points).

Still, it is worth considering two alternate scenarios in the six states listed in Table 2:

  1. The voters disliking both Clinton and Trump give the same support to “other” candidates, but split the two-party votes EVENLY between Clinton and Trump.
  2. The distribution of the “pox on both houses” vote in these six states matches the nation (30% Clinton, 47% Trump, 23% Other)

Table 3 lists how each state would have voted under both scenarios, with the state winner in bold italics.

Table 3: Statewide Vote Distributions in Six States, 2016, Under Three Methods of Splitting Votes of Clinton-Trump Disapprovers

State Actual 2016 results 2-party vote split even Votes split 30-47-23
Clinton Trump Clinton Trump Clinton Trump
MI 47.0% 47.2% 49.1% 45.1% 47.2% 46.6%
PA 47.5% 48.2% 49.6% 46.0% 47.3% 46.6%
WI 46.5% 47.2% 50.5% 43.1% 48.0% 44.4%
FL 47.4% 48.6% 50.0% 46.0% 48.2% 46.6%
AZ 44.6% 48.1% 46.1% 46.5% 45.1% 48.6%
NC 46.2% 49.8% 49.1% 46.9% 46.8% 47.4%

Under both scenarios, Clinton would not only have won Michigan, Pennsylvania and Wisconsin (giving her 278 EV, 8 more than necessary), she also would have won Florida’s 27 EV, for a total of 305 EV. North Carolina’s 15 EV would also have gone to Clinton if the voters who disapproved of her and Trump had split their two-party votes evenly. Arizona, because of its relatively high 7.3% of the vote for neither Clinton nor Trump, would still have gone to Trump under both scenarios.

In other words, the 2016 U.S. presidential was an eight-year/non-incumbent election featuring two historically unpopular candidates. Neither major party candidate had a net positive favorable rating, resulting in an unusually high 18% of the electorate disliking both. Given that this was a change election (net -31% felt country on wrong track), it is not surprising in retrospect that this key bloc of voters chose the Republican nominee (the nominee of the party not occupying the White House), propelling him to the White House.

Still, had the Democratic presidential nominee been viewed even a little more favorably, she might easily have won four additional states with a combined 73 EV, thus winning the White House.

And here is where, if one were to squint hard enough, one could construct an argument that looked something like this:

There is evidence from the RCP averages that Clinton’s net favorability—which was roughly even in June 2015, just as the 2016 U.S. presidential election was beginning—steadily worsened after that, landing at 13 percentage points unfavorable by November 2016. Trump’s net unfavorability, meanwhile, hardly changed over this same period. This could be seen as evidence that Russian interference had the effect of slowly increasing her net unfavorability, to the point where voters nearly disapproved equally of both candidates (then opted for the nominee of the party not occupying the White House).

While this is…plausible, there is one profound flaw (other than the simple fact of NOT explaining why voters who disapproved of both Clinton and Trump then voted heavily for Trump). On January 23, 2013, Clinton was viewed favorably by 63% of American voters and unfavorably by 28%, for a net favorability of 35 percentage points. She had just stepped down from her perch as a popular Secretary of State and was publicly undecided about her future in electoral politics. Still, from that day forward, her net favorability declined steadily and inexorably to nearly even in the spring and summer of 2015.

That is, Clinton was becoming more unpopular long before ANY Russian interference in the 2016 U.S. presidential election. Moreover, her net unfavorability actually hit its nadir (18% net unfavorable) in late May 2016. After that, while the percentage disapproving of Clinton changed little, the percentage approving of her steadily increased.

To me, the bottom line is this:

Democrats are best served understanding that 2016 was a change election featuring two historically unpopular major party nominees. Under those circumstances, an unpopular nominee of the party not occupying the White House is almost certain to beat an unpopular nominee of the party occupying the White House. Period.

Focusing on Russian interference and/or voter suppression as the “causes” of Clinton’s defeat is a wild goose chase. Both are antithetical to a well-functioning, mature democracy and need to be investigated and prevented to the maximum extent, but they also distract from the fact that 46% of the American electorate were predisposed to accept Trump’s message.

Democrats should also realize that Clinton actually defied recent presidential election history by winning the popular vote by just over two percentage points, and that there are strong reasons for optimism in 2018 and 2020 given their growing strength with white college-educated voters, especially women.

In other words, Chill!

Until next time…

[1] Technically, 304-227, as seven Electors voted for other candidates.

[2] In 2016, for example, 12% of self-identified Independents voted for a non-major-party candidate, as opposed to just  3% of self-identified Democrats and 4% of self-identified Republicans.

[3] Given that nearly every poll included in the final RCP averages was of “likely voters,” pollsters did a very good job modeling the actual electorate. This is also indirect evidence that voter suppression did not, in fact, keep an electorally-significant number of Democratic voters from the polls: the projected electorate looked like the actual electorate.

[4] I base this assertion on 1) the fact that voting preferences of voters with an unfavorable view of both major-party candidates had not been assessed prior to 2016 and 2) the historic unpopularity of Clinton and Trump.

[5] Excluding Arizona yields a Clinton 27%, Trump 58%, Other 16% split among the 18% of voters disliking both Clinton and Trump

[6] The fact that Clinton won the popular vote by 2.1 percentage points is even more remarkable in this context, while her 77 EV vote loss was about in line with expectations (22-90 EV loss).

As I head to the APHA meeting in Atlanta in November…

There have been times, especially lately, that I start to write one post and end up writing an entirely different post.

I originally conceived this post to be a simple repository for a set of documents related to my previous career. The impetus for this was two oral presentations I will be delivering in Atlanta on November 7 and 8, 2017.

As I began to explain why I was posting these documents, however, I found myself plummeting down a rabbit hole, describing a series of unpleasant interactions I had with my doctoral committee a few months after I successfully defended my doctoral dissertation in epidemiology.

It made sense to me at the time (doesn’t it always?), but it soon dawned on me that the tone of that section was…off, and that this is simply not the venue to rehash these private interactions, even as I am still processing them.

But once I stepped back (metaphorically, as I was sitting down at the time), I understood more clearly what I was trying to say.

Let me start at the beginning, if you will just bear with me…


While writing my doctoral dissertation, the members of my doctoral committee and I agreed in principle that after my defense we would work together to publish as many as three peer-reviewed journal articles from it (publication was not a graduation requirement).

From my perspective—a 48-year-old married father of two who was 18 years into career as a health-related data analyst/project manager—publication was more “cherry on top” than  necessity, and perhaps also a courtesy to the members of my doctoral committee and other Boston University School of Public Health (BUSPH) personnel to whom I felt grateful.

I defended my dissertation on December 16, 2014. I was not actually in dark shadows, nor was there a bottle of champagne in front of me, but I love this noir-tinted photograph, and it gives you the flavor of that happy day.


This was my moment of vindication, the culmination of a journey I had started 26 years earlier. In September 1989, I enrolled in a doctoral program in government at Harvard’s Graduate School of Arts and Sciences (GSAS). Six years later, I resigned from that program with no degree to show for my time there[1]. But just 15 months later I landed the data analyst gig with a Boston non-profit specializing in substance use and abuse that launched my career. Nine years after that, following a four-year sojourn in Philadelphia, I was back in Boston, enrolling in the BUSPH biostatistics master’s degree program. Four years later, I enrolled in their doctoral program in epidemiology.


I have written elsewhere about the deliberations that led me to walk away from that analytic career towards a writing career (although this blog still allows me to analyze data and write about my findings). That transition “officially” occurred in late June 2017.

However, in February 2017, before I made the career-change leap, I was still actively pursuing positions related to my doctoral studies (assessing the health impact of the built environment, as I detail here).

A few months earlier, I had renewed my long-lapsed membership in the American Public Health Association (APHA); that is how I knew that they would be holding their Annual Meeting & Expo (Meeting) in Atlanta, Georgia November 4-8, 2017. I had delivered work-related talks at their 2001, 2002 and 2003 Meetings, and I had presented a poster at their 2004 Meeting, but I had not attended a Meeting since 2006.

Given that this year’s APHA Meeting theme is “Creating the Healthiest Nation: Climate Changes Health,” it appeared to be a perfect opportunity to advance the job search ball down the field. I thus submitted three abstracts, one for each of my three doctoral dissertation studies. To my surprise, two of them were accepted for oral presentation[2]. And as Meatloaf once sang, “two out of three ain’t bad.”

A few weeks ago, I began to pare the hour-plus-long PowerPoint presentation I had delivered at my doctoral defense down to two 12-minute-long talks. This meant  leaving out many interesting “sensitivity” analyses, including estimates of what my incident rate (IRR) and risk ratios (RR) would have been without exposure or outcome misclassification.

(For a rough translation of that last bit, please see here.)

Realizing how much important detail I was forced to remove from these PowerPoint presentations, I hit upon the idea of making all of the background materials (i.e., my actual dissertation and the PowerPoint defense presentation) publicly available.

And thus you find here:

  1. A PDF of the full text of my doctoral dissertation—Measures of Neighborhood Walkability and Their Association with Diabetes and Depressive Symptoms in Black Women—minus the Acknowledgments (to protect privacy) and CV[3].

Berger Doctoral Dissertation Dec 2014

  1. The PowerPoint presentation I delivered in defense of my dissertation (excluding the “thank you” slides). The last slide was originally this short clip showing the 10th Doctor towards the end of the 2005 episode “The Christmas Invasion.”

Berger Doctoral Defense 2014

  1. The PowerPoint presentations I will be delivering at the APHA Meeting (although not until after I have presented them on November 7 and November 8).

Matthew Berger Measurement Talk 11-7-2017

Matthew Berger Depression Talk 11-8-2017

But this begs a question.

Why haven’t I already published these studies in peer-reviewed epidemiology journals? Isn’t that the usual procedure?

And here we find the rabbit hole I found myself hurtling down as I wrote an earlier draft of this post.


A few months after my successful defense (and once the final logistical requirements had been completed), I received an e-mail from a committee member asking, in effect, where the drafts of my articles were.

Technically, my doctoral dissertation was on track to be published in the ProQuest Dissertation and Theses Global database, where it currently resides.

That is not the same, however, as advancing science through a peer-reviewed publication process; I understood (and had a very high regard for) that then, and I still do now.

But in the spring of 2015, I was still wicked burned out from completing the doctorate itself (with all that had preceded it) while working full time and helping to raise a young family.

I also had higher priorities in my life at that time. My grant-funded Data Manager position was ending in June 2015, and I needed to a) complete the data analysis and final report for that project and b) search for a new gig (or so I thought at the time). My eldest daughter had her tonsils removed and needed a lot of parental TLC. And so forth.

In short, while I was perfectly happy to draft peer-reviewed journal articles from my three dissertation studies, I was not able to do so at that time.

Cutting right to the chase, the member of my doctoral committee and I engaged in an increasingly unpleasant e-mail exchange which ultimately ended in December 2015, when they decided no longer to pursue publication. The details of that exchange are irrelevant.

It is only now, however, that I understand what was really happening then.

For example, as I concluded my Data Analyst requirements, I was actively discussing a related, higher-level position with a different organization. Something kept holding me back, however, and I kept offering (sensible to me at the time) objections. I clearly never accepted that position.

Over the next two-plus years, as I applied to the few relevant positions I could find (58, although some of them were re-postings), my heart was simply never in the search. When I earned in-person interviews, I attended them with what you might call “subdued enthusiasm.” There was always some reason why this position was not quite right…even the last one, in March 2017, that seemed perfect when I first applied.

Even when I was twice offered exciting adjunct teaching positions (I would love to teach again), I ultimately talked myself out of both of them.

Do you see a pattern here?

What I have come to understand as I prepare for APHA, leading me to “publish” my doctoral dissertation here, is that my decision to change careers did not happen a few months ago. It happened, ironically, almost as soon as I walked out of that small meeting room on Albany Street in Boston on December 16, 2014.

In the perceived necessity to find a new position in my then-current career, supplemented with my newly-minted PhD, I could not comprehend, or accept, or grasp, that decision for another two-and-a-half years.

And so this post is not about reliving my unsettling communications with the members of my doctoral committee. It is about squaring a circle, or closing a loop, or whatever “completion” metaphor you prefer.

When I submitted those three abstracts to APHA in February, I was filled with optimism that the November Meeting in Atlanta would be just the place to rekindle my health-related data analysis spark, and where I would joyously engage in the networking necessary to land my next (first?) epidemiology-related position.

It turns out that it will actually be the last hurrah, the period at the end of a nearly 21-year-long sentence.

If you attend the APHA conference next week, I would be thrilled to have you listen to either or both of my presentations.

Otherwise….until next time…

[1] Upon completing my epidemiology doctorate, I finally (and successfully) applied to Harvard GSAS for the Master’s Degree I had earned before resigning.

[2] The incident diabetes study was not accepted.

[3] And, as far as I am concerned, this is tantamount to publication. Consider this passage from the BUSPH Epidemiology Doctoral Program Guidelines (2007, pg. 8): The research…must meet the current standards of publication quality in refereed journals such as American Journal of Epidemiology, American Journal of Public Health, Annals of Epidemiology, Epidemiology, International Journal of Epidemiology, Journal of the American Medical Association, and New England Journal of Medicine. It is understood that the thesis papers may be longer and have more tables and figures than permitted in published papers. Basically, once the members of my doctoral committee signed off on my doctoral dissertation, they were admitting that it already met those standards. Ergo