Big Data

Spreadsheet: Root cause analysis of urination, March 2006 to May 2nd 2009   

Hypothesis: Urination is primarily an intentional, sometimes vindictive, squid-like response. 

Now we'll get down to the bottom of this, thought Rick Pernumero, a data analyst for IBM. If there was one thing his professional life had taught him was that, with a root cause analysis, he could determine the underlying causes of his child's urination and put a stop to it. He resolved to accumulate the necessary data in an incident report until he had gathered a statistically significant volume of data points. 


Date: March 12
Incident: Asked child to buck up.  

Future incidents were not limited to states of buckled or unbuckled, but also over the quality of buckling, such as around the neck not being a valid state of being buckled. 

Date: March 12
Incident: First day driving to new school

The child immediately wet her pants when he announced the trip right before they were to leave. Furthermore, for the duration of the trip, the child flailed their feet around the gear shift and pulled his hair while they were en route to the destination. 

Rick recorded several more similar incidents related to trips to other schools, medical visits, and other daily errands. Despite several near death experiences, He was confident that he was slowly putting another piece of the puzzle together.  

Date: March 19
Incident: Asked to complete an assigned chore: sort through the recycling for curbside pick-up  

Structure was important, he had been told by numerous mental health professionals and community resources, so he did his best to implement a regimen of chores and other routine activities.  

However, child would offer to sweep the floors when asked to vacuum the rugs. When the offer to vacuum the rugs was accepted, child would express a preference for sweeping the floors. When offer to sweep the floors was accepted, child thought it might be better to vacuum rugs. 

Still, he felt he was making great progress. Although no other assigned chores were completed that didn't result in a physical fight, recycling was completed 80% of the time and only 23% of the time with a urination incident!    

Date: April 15
Incident: Asked child to simplify eight sixths

Efforts to do homework were abandoned, but Rick felt that he had accrued an adequately sized homework urination data sample

Date: April 17
Incident: Asked child to give Dad a hug

Date: April 21
Incident: Asked child to eat carrots

Urination incidents appeared to be food-specific. Pizza, strawberries, yams, spinach, and ice cream did not trigger urination incidents. 

Date: April 22
Incident: Asked child if they had wiped their butt with the towel?” 

Date: April 23
Incident: Asked child to brush teeth

This request was discontinued after multiple toothbrushes disappeared. Presumed dropped down heating ducts or buried in the backyard. 

Date: April 25
Incident: Asked child to take medication

Date: April 27
Incident: “Ole Plaid Jacket” 


Date: May 2
Incident: Admonished child for getting silly putty all over the chair

Date: Just about every damned day
Incident: Asked child if they needed to use the bathroom

Date: Just about every time every time I left my dinner unattended and returned to find that it had been eaten.
Incident: Asked child what the fuck happened to my dinner

Date: May 23
Incident: Asked child Why is there fucking coffee in my laptop? 

After 3 months, Rick Pernumero compiled his data and ran it though statistical software and applied factor, cluster, and principal component analyses, regressions, and other analytical techniques. And because he could not concede that there was no discernible pattern in the data and the child's behavior was as random as a roll of the dice, he concluded that he didn't have a sufficient data sample and decided to gather more and more until he could finally see a light at the end of the tunnel.