Sometimes an idea needs to germinate. Last week I mused
about the perpetual uncertainty that parents face in understanding what things
matter when it comes to raising their children. Unfortunately, I feel like I
didn’t get any farther than saying that successes and failures come from some
mix of biological capacity and cultural training and that it’s hard to know
which matters more in a given situation. I did start to talk about the
emergence of advanced metrics in the NBA but only used that to whine that no
such metrics are possible for parents. What I really should have done was to build
on the comparison and imagine what such a system might look like anyway. So,
that’s what I want to do now.
To start
with let me back up and talk briefly again about the NBA system. Owned and run
by a company called STATS, LLC, SportVU is a system of motion tracking cameras and data collection software that was fully installed in all NBA arenas for the 2013-2014 season. It creates a data set 25 times a second, allowing for a host
of experimentation and development in the analysis of player activity on the
court. While the development of new metrics in baseball – where each play can be
broken down into discrete moments of individual input-output measures - took
off back in the 1990s, it wasn’t until the development of this technology
enabled an analysis of holistic secondary and tertiary effects that such
metrics began emerging in basketball.
What I’ve
been playfully imagining over the past several days is something akin to
ParentVU wherein the secondary and tertiary effects of parenting activity could
be analyzed and accounted for. While
plenty of studies have attempted to correlate discrete parental actions with
various measures of child success – for example, time spent reading to children
or meals eaten together – much like the rough measures of points scored and
rebounds gained, these measures don’t really tell us much about how these
things happen and consequently can only give us rough feedback about what
parenting strategies are most effective. A good player scores a lot of points
and grabs a lot of rebounds but that doesn’t give a team a good sense of how
that player fits together with other teammates most effectively. Similarly, a
good parent reads a lot to their children and eats meals with them but that
doesn’t amount to much of a strategic guide when it comes to the full array of
challenges that come with being a parent.
What one
would hypothetically want to do instead is, like an NBA team, record every move
made by a parent and child over time and look for ways to correlate parental
inputs with filial outputs. This, of
course, would require a level of surveillance that feels downright terrifying
on the face of it, but for the sake of this post I’m going to skip past the privacy/ethical/logistical
quandaries for now and imagine what kind of metrics we might be able to develop
from it. I’ve come up with several from my own experience that I’d love to
track.
- “Do as I
say” ratio – tracks the consistency between what parents ask children to do and
the things parents do themselves
- Eye
contact percentage – tracks quantity of eye contact during time parents and
children are around each other. Would be a proxy for how attentive parents are
to children in various contexts – the playground, the dinner table, etc
- Touch
matrix – tracks the number and types of physical contact between parents and
children as well as categorizing them into negative and positive groups. The
number would be some kind of ratio per hour and positive/negative ratio.
-
Consistency index – tracks similar situations over time and determines the
consistency of parental response. This would make no value judgments about the
quality or nature of these responses. It is merely looking for whether
responses are the same or different over time.
-
Conversational exchange rate – tracks number of conversational exchanges versus
other exchanges such as corrections, instructions, punishments, etc.
As I put this list together a
couple of caveats came to mind. One is that I am imagining metrics from a
parental experience point of view and not a quantitative measures point of
view. There are probably all kinds of interesting twists to be measured by
someone better versed in the math of these things than I am. Second, this list
reflects a very particular, very personal version of what makes a good parent.
There are no metrics listed above for ‘one hit wonders’ such as buying
spectacular presents or undertaking big displays of affection. To me these
things smack of parental self-aggrandizement and I don’t think statistical
tracking of such things isn’t going to lead to any great revelations about
parenting strategy. Third, the metrics also reflect the current age of my
children (6 and 8) and might be somewhat different if they were much younger or
older. Fourth, I’m favoring percentage and ratio measures over cumulative ones.
While something like number of diapers changed or soccer games attended could
tell a good deal about the time a parent spends with a child, such measures
would also get skewed by the different roles parents have in the household and could
also tend to mistake quantity for quality. I’m less interested in how much time
a parent spends around their children and more curious about what they’re doing
with the time they do spend.
Another
interesting challenge – and one that I’ve thought less about – is how to
measure results and draw correlations between them. This requires its own set
of value choices regarding what kinds of personalities one wants their children
to have, which strengths one values, which weaknesses one can live with. Then
too, parenting is so much about investing ahead of time to produce results down
the road that connecting the dots between input and output would be especially
difficult. With all of that said, here are a few possible output metrics that I
would love to connect to my parental input variables:
- self
confidence
- respect
for others
-
willingness to listen and incorporate the ideas of others
- ability to follow directions
The data provided by SportVU has
changed a great deal about the strategies NBA teams employ in this day and age.
After analyzing shot percentages and per possession scoring data, teams now
work especially hard to create shots within 8 feet of the basket and beyond the
three point line, disdaining longer two point jumpers as inefficient and less
likely to produce positive results. Fluid passing and the management of court
space have become more important than ever as teams look to get defenses
scrambling around before they dump the ball down low or out to high percentage
spots around the three point line. In turn, defenders who can quickly drop into
the paint and then jump back out at shooters have become high valued as well.
While a
comparative ParentVU system would be complex beyond measure, it’s fun to
speculate on what new strategies for improvement might come from them,
particularly if the data could be collectively analyzed and shared across a
large population. Would parents check their stats the way they do their Fitbits
to see how well they’re doing on the various metrics? Would they compete to see
who had the best touch matrix for the month? Would such a system spawn a whole
new approach to parenting akin to the NBA’s embrace of ‘pace and space?’ Would
that actually give parents a better sense of which things they can do to
positively mold their kids?