Archive for November, 2016

How a Configurational Contextual Model ‘Trumped’ the Conventional Forecasters: Four I’s Proved to be Powerful Predictors in the 2016 Presidential Election

by Steve Montague

Groups like Pollyvote and Real Clear have been doing a fair bit of ex poste diagnosis of which models and approaches did worse than others in failing to predict the 2016 Donald Trump US Presidential election victory.  This article will not rehash those arguments other than to note that all sources have had to admit that what they have called a ‘configurational’ or ‘threshold’ model developed by a history professor name Lichtman outperformed the more elegant statistical and econometric models on offer.[1]

This has to be seen as a bit of a coup for analysts and evaluators who believe that causation can be thought of as the product of contextual factors playing on any mechanism or set of mechanisms at a given period of time.

The Lichtman Keys can be represented as statements that favor victory for the incumbent party. According to the model, when five or fewer statements are false, the incumbent party is predicted to win the popular vote; when six or more are false, the challenging party is predicted to win the popular vote.

  1. Party Mandate: After the midterm elections, the incumbent party holds more seats in the U.S. House of Representatives than after the previous midterm elections.
  2. Contest: There is no serious contest for the incumbent party nomination.
  3. Incumbency: The incumbent party candidate is the sitting president.
  4. Third party: There is no significant third party or independent campaign.
  5. Short term economy: The economy is not in recession during the election campaign.
  6. Long term economy: Real per capita economic growth during the term equals or exceeds mean growth during the previous two terms.
  7. Policy change: The incumbent administration effects major changes in national policy.
  8. Social unrest: There is no sustained social unrest during the term.
  9. Scandal: The incumbent administration is untainted by major scandal.
  10. Foreign/military failure: The incumbent administration suffers no major failure in foreign or military affairs.
  11. Foreign/military success: The incumbent administration achieves a major success in foreign or military affairs.
  12. Incumbent charisma: The incumbent party candidate is charismatic or a national hero.
  13. Challenger charisma: The challenging party candidate is not charismatic or a national hero.

For each of these Lichtman constructs what amounts to a ‘truth table’ (see Qualitative Comparative Analysis for an explanation of Truth Tables) which is essentially a set of either true or false (1,0) ratings – along with a certainty level – for each of the  13 factors. The evolution of these ratings over time leading up to the November 8th 2016 US election is a matter of public record. Lichtman now famously predicted a Trump victory as early as September.

What may be less obvious is the way that the Lichtman factors seem to cover what Pawson has called the four Is of context. These four Is include Infrastructural considerations, Institutional considerations Inter-relational considerations and Individual considerations[2]. Thinking about these categorizations is useful here because there has been a tendency to simplify Lichtman’s findings into concluding that the fate of an incumbent administration is completely ‘up to them’. [3] When you look at the four Is categories – you can see that the Lichtman ‘keys’ each fit in to at least one of the four I’s. See below:

1: Party Mandate Infrastructural
2: Contested Nomination Institutional
3: Incumbent Status Infrastructural
4: Third Party Challenge Infrastructural
5: Short-term Economy Infrastructural
6: Long-term Economy Infrastructural
7: National Policy Achievement/Shift Institutional
8: Social Unrest Infrastructural
9: Scandal Inter-relational / Individual
10: Foreign Policy Defeats Inter-relational
11: Foreign Policy Success Inter-relational
12: Incumbent Charisma Individual
13: Challenger Charisma Individual


These ratings are mine and certainly some could be disputed or rendered into more than one category. The important thing though – is that the range of contextual factors goes from the broad socio-economic and political – products of history and broad circumstances, through to decisions largely made by key institutional groups (like national policy shifts), through to inter-relational factors like foreign affairs policy wins or defeats through to individual characteristics – like the charisma of the incumbent and challenger leaders. So in fact, many of the factors influencing the selection of president are out of the hands of either the incumbent or the challenger.

When it comes to strategy then – the idea is to focus on things you can most easily influence.  If you are a challenger the implication is that one should focus on the factors which could be most easily interpreted as associated with a person and their relationships. Note that some of the main messages of the Trump campaign focused on factor 9. scandal (Clinton’s use of a non-government sanctioned email account) factor 10. foreign policy (in terms of trade deals being ‘horrible’, in addition to being soft on terror etc.) and factor 12. incumbent charisma/character (characterized by Trump as ‘Crooked Hillary’ and the accusation that she was somehow ‘low energy’).

So the Lichtman model prediction success should suggest two important things for analysts, researchers and evaluators:

  1. Context is critical to understanding outcomes; and,
  2. A configurational ‘cause-effect’ model (possibly ‘sorted’ by the four Is as shown here or by some other kind of contextual leveling which range from the broad to the specific) can help to explain (and predict) results.

In our recent work we have been working on what might be called check lists for factors that affect success in terms of policy and program implementation. For a related article see It appears that this latest development validates the further pursuit of such configurational approaches in various fields of application.  

Steve Montague ( is a partner with PMN, a Fellow of the Canadian Evaluation Society and an adjunct professor at Carleton University in Ottawa, Canada.

[1]Lichtman, Allan J. (2008). The Keys to the White House: A Surefire Guide to Predicting the Next President (2008 ed.) New York, NY: Rowman and Littlefield Publishers The approach has predicted every Presidential election since 1984.

[2] Pawson, R. and Tilley, N. (1997) Realistic Evaluation Sage  Pawson outlines a “four Is” framework as follows: Infrastructure (which refers to the wider social, economic, and cultural setting of a program/intervention); Institutional setting (the characteristics of the institution involved); Interpersonal relations (nature and history of key relationships); and, Individuals (characteristics and capacities of stakeholders).

[3] Lipman himself says this as follows “The principal historical lesson to be drawn from the Keys is that the fate of an incumbent administration rests largely in its own hands; there is little that the challenging party can do to affect the outcome of an election.” Lichtman, Allan J. The Keys to the Whitehouse op cite  

Impact Pathways for Science Initiatives Released

A well received report describing science impact pathways was recently noted by an NRC official at the November 7th Science-based Organization public forum.  The report Study of Large Scale Research Infrastructure Impact Assessment was co-authored by PMN partner Steve Montague and associate Gretchen Jordan.  The full report has just been released for public access.  Email to receive a copy of the full report .