Page 104 - MYM 2015

those questions, often by looking at short term results, i.e., intent to purchase, preference and the like.
(Kim, 2011) While these short-term impact factors
are clearly important to marketers, the longer term implications would seem more valuable for such things as identifying long-term trends, development of brands and branding policies and the like.
To investigate this question, data from a continuing Prosper study was used. Using responses to
monthly questions posed in an American Pulse-type study, Prosper researchers set up what they call a “Happiness Index”. Similar to the measurement of consumer sentiments which are measured by a number of organizations, (Prosper Insights and Analytics, Goprosper.com) this index was created to determine the emotional quotient of the U.S. consumers on an ongoing basis. Using a scoring format, respondents
are asked to say how “Happy” they are with ten major factors in their lives, i.e., (1) house/apartment/condo, (2) love life, (3) home life, (4) work life, (5) neighborhood where they lived, (6) relationships with family, (7) relationships with friends (8) health, (9) government, and (10) religion/faith. Prosper and their clients use these measures to forecast such things as likelihood of moving, changing jobs, health concerns, etc. We use that same data but in a somewhat different way, i.e., to forecast and classify respondents and investigate how they likely will or will not change as they move through the aging process. While we should note that these factors are highly inter-correlated, they do provide some insights into what occurs generation to generation and as people move from one life / age stage to another.
We start with a discriminant analysis of the data from the November, 2014 measurement which consisted of 6,593 respondents. Exhibit 12 shows the result of that exercise.
The question next was could age groupings or cohorts add some predictive value to this initial identi cation of the data groupings. Exhibit 13 shows how the age groups relate to each other when plotted on the Love
Exhibit 13
Happiness Functions Predicting Ages
25.34
0.2 0.15 0.1 35.44 0.05
L O
VE
&
H O M E
65+
-0.4
18.24
-0.2 00 -0.05
-0.1
-0.15
-0.2
0.2 0.4 0.6 55.64
45.54
HOUSE
0.8
Exhibit 12
Happiness Structure Matrix Monthly Survey - November 2014
Happiness with... 1 2 3 4 5
House/Apartment/Condo .60*
Love life .84*
Home life .72*
Work life .62*
Neighborhood
Religion/faith
.53*
and Health Happiness (Y axis) compared to Housing Happiness (X axis). (Note unlike other plots of similar data, the size of the dots in this chart is locational only. It does not indicate the size of the respondent base. Also, only two of the functions are plotted on this chart, simply to reduce the information overload).
What this chart shows is a prediction of the age when each of the two Happiness Functions are most likely to occur. Clearly, age has much to do with the happiness people have with their home and housing. The older they get, the more settled they are and seemingly the happier they are with their surroundings. Alternatively, those who are less happy with their housing are found in the 18-24 group, while the 25-34 group is only a bit happier, while those in the 35-44 group seem to be moving forward in their housing happiness.
When we look at the Love/Relationships plot, clearly the 18-24 groups are the least happy with their situation. In fact, it is this Millennium or Digital group which seems the least happy of all the various age groups. In spite of this age group’s aggressive social life, as characterized in the media and advertising, this age cohort appears
to be the least happy with both of the two factors. While these results are preliminary and exploratory, they
do appear to explain the less settled, more transitory nature of this age group. The encouraging thing, however, is that this type of big data does seem to suggest that it might well be possible to forecast and/ or predict the impact and effect of any number of social and behavioral factors and the stages these age groups might pass through. That could be most helpful to the long-term planning by marketers and researchers.
Based on this preliminary analysis, it seems safe to propose the next research postulate.
Relationships with family .50*
Relationships with family .55*
Health .53*
Government .61*
As can be seen, that analysis created ve groupings. The rst has to do with house and home, the second with relationships, the third with friends and health, the fourth with government and the fth with religion or faith. As can be seen, the elements with the highest loadings are love life and home life. Not a surprising nding but, one that seems to ground the entire study.
104 I October 2015
.52*