Psychometrics predicting behaviour, in theory and in practice
This two-part talk, the first will look at how personality influences behaviour in social media, tracing the limits and analysis and prediction and comparing to behavioural and keyword targeting. The second part will take a practical look at how data this type of technique can be used to obtain psychological insights and targeting strategies for any key brands or audiences.
Find more info at: http://chinwag.com/insight/psychology
What Marketing Can Learn From Vanilla Icepeteraharris
Ähnlich wie 'Psychometrics Predicting Behaviour, In Theory and In Practice' Dr David Stillwell & Dr Stephen Haggard, Cambridge Personality Research (20)
'Psychometrics Predicting Behaviour, In Theory and In Practice' Dr David Stillwell & Dr Stephen Haggard, Cambridge Personality Research
1. David Stillwell Stephen Haggard
david@mypersonality.org stephen@mypersonality.org
Science Director Business Director
Research Associate – The
Psychometrics Centre,
Cambridge University
2. myPersonality
• Running since June 2007
• Allows Facebook users to take
real psychological
questionnaires and receive
feedback on their results
• Have data for ~6.5m users
• Used by hundreds of people
every day
• 25+ reliable and valid
questionnaires available
• Users can opt in to sharing
their FB profile data
https://apps.facebook.com/mypersonality
6. Films liked by Ages 30+
Introverts Male Female
•Firefly •Girl, Interrupted
•Star Wars •Labyrinth
•The Last Samurai •Love Actually
•Sin City •16 Candles
•Kill Bill •Eat Pray Love
•Pulp Fiction •Grease
•Dumb and Dumber •Dirty Dancing
•Super Troopers •Madea
Extraverts
7. Psychology-based working gives
leverage in social network missions
Application: Facebook ad targeting
stable patterns emerge early in campaigns
accessible through standard inputs eg keywords, adserving
responds to standard techniques eg optimise
blend with other approaches behavioural, social-dem, creative
8. Case Study: Fanning Campaign
• Client : insurance retail brand: student’s room
insurance product
• Target: 60mn impressions to 16-17 y.o. pre-
college FB Users
• Method: display ads promoting FB “Win £500”
competition
• Planning: half of spend is on personality-based
target groups
9. Case Study: fanning campaign
• Target: 60mn
impressions to 16-17
y.o. pre-college
• Method: display ads
Client : insurance retail brand for “Win £500”
Product: student room insurance competition
• Planning: half of
spend is on
psychologically
defined target groups
10. Success: conversions
A leading Agency’s Facebook competition-entry campaign compared its in-house keyword generation system
against personality-based keyword lists generated by Cambridge Personality Research Preference Tool in a full
campaign of 42 mn impressions in November 2011. The top three performing segments of each method are
compared. Taking conversion as the metric, target groups defined by Personality using the Preference Tool
outperformed target groups defined by the Agency by an average of 45%. Note that CPM was, however, 30 %
higher on average for these top-performing groups.
11. the downside: spreads
• The spreads on conversion are wider
• Psychology segments range: 2 – 35
• Agency segments range: 5 – 19
• Agencies could “game” the system
• “Likers” psychodynamic – best CTRs
• But client needs conscientious people
- whose CTR is only marginally better than the
best agency segment
12. BRAND INSIGHT & CREATIVE
New Areas
Language, tone in community content
Creative & Media personalisation by
psychology
Brand Insight Reports
Available Free to Chinwag delegates
13. David Stillwell Stephen Haggard
david@mypersonality.org stephen@mypersonality.org
Science Director Business Director
Research Associate – The
Psychometrics Centre,
Cambridge University
Hinweis der Redaktion
16PF, MBTI with 4 factors The first major inquiry into the lexical hypothesis was made by Sir Francis Galton . [40] This is the idea that the most salient and socially relevant personality differences in people’s lives will eventually become encoded into language. The hypothesis further suggests that by sampling language, it is possible to derive a comprehensive taxonomy of human personality traits. In 1936, Gordon Allport and H. S. Odbert put this hypothesis into practice. [41] They worked through two of the most comprehensive dictionaries of the English language available at the time and extracted 17,953 personality-describing words. They then reduced this gigantic list to 4,504 adjectives which they believed were descriptive of observable and relatively permanent traits.