What defines us? What makes up our ‘identity’?
As part of its large (3594 person) 2014 British Values Survey, CDSM (Cultural Dynamics Strategy and Marketing) asked two ‘my identity’ questions, with 31 ‘facts’ offered as options important in forming identity. The results are published for the first time in this post.
In one question Cultural Dynamics asked people to pick as many of the 31 ‘facts’ as they liked, and in the other, to select the three they found ‘most important’ (including a ‘none of these’ option). Some charts of the results are presented below, and images of the full data sets can be downloaded here.
The 31 ‘fact’ options were:
My nationality (English, Welsh, etc); Being British; My county or city (Yorkshire, London, etc); My local area; Being a parent; Being European; My social class; Being the sex I am; My skin colour; My religion; My tastes; My occupation; My standard of living, possessions; My family history; My age, stage of life; My intelligence; My creative abilities; My emotions and feelings; My imagination and fantasy; My practical abilities; My political convictions; My educational achievements; My interests; My principles and values; My circle of friends; My income; My body, face, hair; The way I dress; The way I speak; My ethnic origins; None of these.
The survey also collected information to segment the results by Motivational Values (the three Maslow Groups of Settler, Prospector and Pioneer, and the 12 Values Modes within them), and by sex, age and class (Socio Economic Group).
Chris Rose, firstname.lastname@example.org download this post as a pdf here
Above: 31 ‘identity’ factors used in its values-segmented 2014 survey by CDSM (www.cultdyn.co.uk). [This graphic was not part of the survey!]
It hardly needs saying but the more important ‘identity factors’ play a big role in intuitive responses to attempts to communicate with audiences, providing reflexive ‘Track One’ answers to questions such as “is this about me?” or “does this person understand me and my life?”
Some Findings Which Might Interest Campaigners and the ‘Political Classes’
Taking the samples as a whole, the five most frequently chosen ‘facts’ when invited to ‘Choose all the facts you feel are important in your identity – who you feel you are’ were ‘my interests’ (1), ‘my principles and values’ (2), ‘my intelligence’ (3), ‘my nationality’ (meaning English, Welsh, Scottish) (4), and ‘My emotions and feelings (5).
The five most frequent when asked to ‘Choose the THREE facts that are MOST important to you’, were: ‘my principles and values’ (1), ‘being a parent’ (2), ‘my intelligence’ (3), ‘Being British (4), and ‘my emotions and feelings’ (5).
In some cases there are quite marked differences in the choices in relation to values, age, sex, or class (later), which may be relevant to audience targeting. In other cases there are no such differences, meaning that these are potential options to reach ‘across divides’.
Overall, the two ways of asking people to chose between the options gave similar results (above). The 13 most frequent choices are the same in both cases, although the order is slightly different. (In most of this blog I focus on the ‘top three’ results as that question forces people to think about their response a bit more and so gives greater discrimination eg across values. But in some cases users don’t need or want maximum discrimination but to see even weak effects. Readers can find the full data here).
A number of options touched on factors frequently debated in the news and social media on identity grounds but many of these do not appear in the more frequent choices.
For example, despite the huge amount of media discussion about sexual identity, politics, and feminism, ‘being the sex I am’ came in (top three question format) at rank 21 (in 3.6% of the choices), ‘my political convictions’ ranked 25th (2.7%), and ‘my ethnic origins’ and ‘my skin colour’ were both included in less than 2% of the ‘top three’ selections. (See table below).
I don’t know if this is encouraging or discouraging to campaigners and policy wonks who spend an awful lot of their professional or activist time (much on ‘Track Two’) on issues of gender or diversity but at least in terms of self-identity, this suggests that as a whole, the British do not often define themselves in these ways. Nor do they often define themselves by ‘social class’, ranked 30th at 1.4%: one for Jeremy Corbyn to ponder on perhaps?
Bottom of the list came ‘being European’. This survey was conducted in October/November 2014, after January 2013 when David Cameron announced there would be a referendum on EU membership but before the European Union Referendum Act 2015 was passed and before he announced the June 2016 referendum date, in February 2016. It is possible that the massive subsequent pre-occupation with ‘Brexit’ may have raised the priority for ‘being European’ but it is very unlikely that it has changed the low rating for the importance of ‘political convictions’, which was also seen in previous versions of this survey. People who spend a lot of time ‘in politics’ or watching politics and ‘issues’ (like me), tend to massively over-estimate the public interest in what they are doing or consider important.
In Britain quite a lot of people vote but very few put ‘my political convictions’ in the top three of their identity factors. ‘Very political’ people are very different from most of the British population.
Values and Identity
Readers familiar with the Values Modes model will know that because it creates groups from how people think, by measuring hundreds of attitudes and beliefs, values groups are in effect already an identity mapping exercise, in that they show sets of correlated convictions about how the world ‘really is’. So for example, people in a particular Maslow Group (Settler, Prospector or Pioneer), or within in a Values Mode, will soon detect whether or not other people are ‘like them’, and in situations where they can exercise free choice, often end up socialising with people in similar groups.
Here’s the 2014/5 British Values Map. Settler is top right, Prospector left, Pioneer lower right.
This shows the 100 ‘Attributes’ which statistically most separate the different values groups. Each can be plotted as a single ‘map’ but here they are shown (the dots) at their points of maximum ‘espousal’, the point on the map where they are ‘strongest’. Behind this map is a 1000×1000 grid of survey responses, in effect like combining the results of a thousand separate surveys.
Links to explanations of the Values Modes system and more of my posts on values can be found here, and at CDSM’s website (including an alphabetical description of the Attributes). See also my book What Makes People Tick, The Three Hidden Worlds of Settlers, Prospectors and Pioneers.
Understanding motivational values, which along with framing and heuristics are major drivers of everyday behaviour on ‘Track One’, gives a much greater insight into social dynamics in which identity plays a part, including politics and events like ‘Brexit’. (See analysis of how that came about here and the insights of before and after referendum values and voting surveys here).
This ‘identity’ survey overtly asks people to think about their identity and offered 31 options, some of which are also used in other identity surveys, allowing for some comparison.
Here’s the overall response table with values skews shown to the right.
The coloured boxes indicate significant positive or negative associations at 95%, 97.5% or 99% levels. Warm colours indicate positive association, in other words that Maslow Group (Pioneer, Prospector or Settler) ‘over indexed’ on selecting that option, compared to the population average response. So for instance Pioneers indexed 127 on ‘my principles’ and values (option ranked 1), 27% more than the population average, and although a lot of Settlers and Prospectors also ‘ticked that box’, Prospectors were 16% less likely to do this than as if ‘by chance’ (index 84), and Settlers (88) were 12% less likely. In contrast, there is no significant difference in values terms in the case of ‘my emotions and feelings’ (ranked 5). The ‘index’ takes into account the different sizes of the three Maslow Groups in the population (this survey found 34.5% Pioneers nationally, 36.9% Prospectors and 28.6% Settlers), as well as the response to the option.
Eight options (below) showed no values difference in responses at the Maslow Group level. Of these ‘my emotions and feelings’ and ‘my circle of friends’ are popular responses, so if you started a conversation or created a proposition assuming these were important to identity on either of these bases, in Britain it would be very unlikely to trigger any values-inspired rejection.
Most of the responses are differentiated by values, and the coloured ‘skews’ are a quick way of identifying these. However looking at the skews alone can mislead us into overlooking the fact that substantial numbers of people from ‘under indexed’ or ‘average’ Groups also chose that option. Here are the raw numbers of respondents from the 10 options most selected as in my ‘top three’.
This shows that the highest frequency of ‘my principles and values’ is down to support from all three Groups but with disproportionate support from the Pioneers. What is meant by ‘my principles and values’ will be very different for each group, although with some things in common between pairs of groups. Any conversation about ‘principles and values’ across Groups could start to diverge almost immediately.
The second most popular choice was ‘being a parent’ and this is also the most evenly matched between the Groups, although it shows as ‘skewed’ to Settler as it is chosen by a disproportionately large number of Settlers. ‘Being a parent’ is more founded in common experiences than ‘principles and values’, and so offering a lot more potential ‘common ground’ and scope for agreement. (Eventually it also would start to diverge, for example on the nature and objectives of ‘good parenting’ and the ‘right’ structure of ‘families’). This is why I often advise communicators in Britain, that good starting point for communications deliberately or by default aimed at a mixture of values groups (eg “the public”), stands a better chance of ‘getting a hearing’ if it framed as about children or parents/families (cf for instance just ‘nature’ – see this example of the effect).
Testing by CDSM has not shown any difference in intelligence based for instance on IQ, between values groups. There are differences in educational level, and although this is a hotly contested topic, it is very likely that this is in part due to social advantages (eg the influence of richer parents), and the effect of the educational process in enabling values-transitions, especially from Prospector to Pioneer (achieving esteem and self esteem).
So when Pioneers over-index on ‘my intelligence’ as an identity factor it is probably not because they are more intelligent but because they value ideas (for instance more than things) and have an unmet need to explore new ideas and connections. From this, they may conclude that ‘intelligence’ is important. It has to be said that one of the more annoying tendencies of Pioneers is to attribute their convictions to having made ‘the right’ (meaning clever) choices, and to have a lot of ‘facts’ and arguments available (as they spend time collecting them) to back these up. This is why the other Maslow Groups often refer to them as ‘smug’.
By the same token, although ‘my interests’ was chosen by a lot of people from all three Groups, the Pioneers over-indexed, and they do tend to have more different ‘interests’ and greater active curiosity. Similar reasons lead Pioneers to score ‘my creative abilities’ highly (whether or not other people think them very creative, it’s often important to them).
Two stand-out Settler over-indexes are on ‘my nationality and ‘being British’. This topic became hugely discussed as a result of the ‘Brexit vote’, and at its simplest, the Settler emphasis on national identity is driven by an unmet need for safety, security and belonging. See for example the discussion on perceived threats to cultural identity from immigration, in The Values Story of the Brexit Split (Part 1) [see slides 44-60 including on the authoritarian response to cultural change].
Brexit split the values map across the middle along a pre-exiting fault-line over ‘Europe’:
(above: attitude to EU, 2015; below, the Leave vote)
Identity was not the only factor but it was an important one. It is interesting that nationality rather than geography and ‘place’ produces the higher results, across all values groups. The option “my county or city, eg Yorkshire … London etc” came in 17th when people were asked to pick the three most important factors (included by 4.2%), and ranked 19th when participants could select a many of the ‘facts’ as they wished (19.1%). Likewise ‘my local area’ ranked 20th (at 3.6%) when people picked their ‘top three’ identity factors, and 16th (22%) in the unrestricted choice.
Writer and editor David Goodhart attracted a lot of ‘Brexit’ comment in 2017 when he proposed in his book The Road to Somewhere that the British now divide into ‘tribes’ of people based on affinity (or lack of it) to place or local cultural continuity: the ‘Anywhere’s’ (liberal, about 25%), ‘Somewheres’ (the reverse and ‘about half’ the population) with a strong connection to place, and ‘Inbetweeners’ (those ‘in between’ – about 25%). This CDSM survey specifically asks about ‘my local county or city’ and ‘my local area’ and neither produce any sort of result suggesting this is a defining identity factor for 50% of the population.
Geographic determinism is a popular option for political pundits in Britain and the US, perhaps because they are two of the few countries with a first-past-the-post electoral system based on geographic constituencies. There was much reference to ‘Northern Towns’, ‘forgotten’ seaside towns and ‘Metropolitian Elites’ and ‘Rustbelts’ in media explanations of the EU Referendum result but the ‘locational’ explanations may owe more to half-remembered school geography books fished from journalistic Pensieves, than any analysis which stands up to scrutiny.
Values analysis produces a better explanation but the social geography of values is far too fine-grained to produce such convenient handles as Anywheres v. Somewheres. It is likely for example, that Settler (and Golden Dreamer and Happy Follower Prospector) attachment to people-and-places-I-know, is real but subsumed in some of the identity response captured in ‘my circle of friends’ and ‘my emotions and feelings’ but that’s not just about geography and where you are ‘from’. Nor, in our mobile and online-connected world, are any of the values groups now confined to making social connections through face to face contact within ‘their local area’.
The different choices of three most important identity factors made within the main values groups may be of use to anyone thinking about how to engage these groups (above). ‘My principles and values’ is a great place to start but requires a lot more insight than ‘being a parent’, while ‘Being British’ is a stronger factor for Settlers and Prospectors than Pioneers.
‘My body, face, hair’ creeps in at 10 for Prospectors: about appearance and looking good. ‘My age, stage of life’ appears at 10 for Settlers, largely due to the cohort effect (it is a more frequent choice for older people and Britain’s current Settler population skew older).
Differences by Values Mode
The full table of twelve Values Modes against 31 options is too big to reproduce here (you can download it here) but below is a table of the ten most popular choices, extracted from the ‘pick three’ responses (showing only the indexes or skews).
In this table I have transposed the Values Modes into their ‘transition order’, from RT (Roots) to TX (Transcender). CDSM research suggests that individuals ‘transition’ from one Values Mode to the next, if they do, along this sequence:
The names given to each Values Mode by CDSM are shown below, together with a schematic version of the ‘Values Map’, also showing the priority need of each of the ‘outside edge’ Values Modes:
This table shows that three of the four Settler Values Modes (VMs) over index on ‘being a parent’, and three of the four Pioneer VMs on ‘my principles and values’. The ‘outside edge’ VMs (see schematic map) are typically those with strongest values identities, and these VMs tend to define and dominate values dynamics (eg change or resistance to it). The TX Transcender VM is frequently wildly over-represented amongst leaders of organisations, particularly those concerned with ‘issues’. (You can take the values questionnaire and find your own Maslow Group and Values Mode from the CDSM website survey tool here).
TXs over-index on ‘my principles and values’, ‘my intelligence’, ‘my interests’ and ‘my creative abilities’ as identity factors, and strongly under index on ‘being British’ and ‘my nationality’, and slightly less so on ‘being a parent’. On the other side of the Values Map, the ‘Brave New World’ and ‘Certainty First’ Settlers show almost the exact opposite skews. This is the ‘Power v Universalism’ axis discussed in several previous blogs including on ‘Brexit’.
A key ‘swing’ group is the NP Now People Prospectors, who can act as a bridge for the spread of new attitudes and behaviours from Pioneers (taking them from the TX) and popularising them with other Prospectors. It is notable that the identity factor ‘being British’ shows over indexes in all Settler and the first two (GD and HF) Prospector VMs but is then under strongly indexed in NP and TP (the similar Tomorrow People). This is the values inflexion across the middle of the values map, which was present in the EU/Brexit divide. But it’s not the case for ‘my nationality’, being English, Welsh or Scottish, on which only the Settler VMs are over indexed. I don’t have a good explanation for why this is.
It may be that the clarity of national identity – a binary in/out, presence or absence quality – acts as a simplifier, giving symbolic certainty which is satisfying to Settlers, whereas ‘my local area’ or ‘my county’ or town/city is harder to see in this way because everyday experience shows it to be more complex and less definitive. I also wonder if ‘British-ness’ exists in juxtaposition to an outside influence (eg supposedly, as in the Boris Johnson caricatures, the EU). But to investigate this would require qualitative research.
Like the Pioneers, the NPs also over index on ‘my intelligence’.
For more on the differences between individual VMs, follow the links on the home page at www.campaignstrategy.org
Above: the overall results in rank order with indexes showing the significant male/female differences. About two thirds show sex differences.
Identity factors chosen by significantly more males:
The strongest over-index is on ‘my political convictions’. Although this is a tiny group, it is a very male-dominated choice. The next strongest skew is on ‘my county or city’. I can’t help wondering if this might have something to do with affinity to sports clubs.
Identity factors chosen by significantly more females:
The biggest difference is on ‘my body, face hair’ (89 points), followed by ‘being a parent’. The latter is most relevant in ‘targeting’ terms as this is a much more popular choice. Together the top three probably illustrate the political or campaign significance of female dominated blogs, websites and media channels covering ‘classic’ “women’s issues”.
‘My principles and values’ is close to gender neutral, and probably is so amongst Pioneers.
Identity factors chosen equally by females and males:
Age and Identity Choices
In this case I have used the ‘chose as many as you like’ question and shown only those with clear age effects.
Those identity factors more important to older people:
The clearest age effect is ‘being a parent’, which also looks like an experience-related effect. In other words’ it’s caused by the real-life experience of having children and being a parent. It is also cited more frequently as people age.
Nationality, being British and ‘my local area’ all show similar age-related increases in frequency, only part of which can be down to the Settler-older correlation.
‘My political convictions’ and ‘My principles and values’ would be interesting to explore with qualitative research. CDSM has made many studies of political affinity and voting in the UK, and shown strong values effects which tend to be quite consistent or slowly changing with respect to Labour and the Conservatives and Settlers and Pioneers but much more labile in relation to Prospectors (typically swing voters). The values profile of UKIP, the Greens and the LibDems is much narrower and more static. It seems this is not the same as ‘political convictions’ as an identity factor.
Those identity factors more important to younger people:
Fewer identity factors are skewed to the young. As with ‘being a parent’ it is tempting to see some of these as lifestyle pre-occupations. For example the salience of ‘my occupation’ falls of a bit of a cliff at 34 just as ‘being a parent’ takes off.
Finally, ‘age and lifestage’ as an identity factor, in relation to age:
This shows a different pattern over-indexing at each end of the spectrum, perhaps because the effects of age when very young and when increasingly old, become things that ‘middle aged’ people rarely have to think about.
Class and Identity
Lastly, we can look at the relationship between ‘identity’ factor choices and ‘class’, which in Britain is conventionally measured by Socio-Economic Group, itself defined by occupation.
The table above shows results from the ‘pick your top three’ question along with skews of significance by Social Class. (AB is ‘professional’, C1 ‘clerical/ supervisory’, C2 ‘skilled manual’, DE ‘unskilled’, ‘unemployed’ and ‘retired’; student here is coded as C1).
There are differences in the British population across values groups and SEG, although they are not strong or consistent enough to treat one as a substitute for the other:
(more recent survey data)
As discussed in Brexit Values Story Part 1 and Brexit Values Story Part 2.1, the broad correlation between approving or not of Europe and voting for Brexit or not, with class and values, is consistent between CDSM values surveys and others such as Lord Ashcroft’s survey. This is obvious in the case of identity factors such as ‘Britishness’ and ‘nationality’ and probably hidden within the responses to ‘my principles and values’.
The over indexes on ‘my intelligence’ and ‘my interests’ amongst ABs are at least partly due to the auto-correlation with Pioneers and ABs. The over indexes amongst ABs on ‘my standard of living, possessions’ and ‘my occupation’ are at least partly due to these also over-indexing with Prospectors (ie ‘successful’ people), who also over index on ‘my educational achievements’.
Acknowledgement: thanks to Les Higgins and Pat Dade of CDSM (email@example.com) for permission to use these data