Lloyd's Angel: Unexplained Phenomena
November 1961
It was ludicrous, but I couldn't tell Dr. Reynolds that. I might have been short-sighted enough to tell him anyway, but my mouth and brain were frozen in stunned surprise until the opportunity was past.
Finally I just picked up the notes and left without saying anything. I was convinced none of this would have happened if Dr. Needum hadn't been on sabbatical, but he was -- and my Ph.D. prospects were in Dr. Reynolds' hands for this academic year.
It was expected that Reynolds would have me doing his scutwork. It was, perhaps, bearable that he had me doing busywork for the benefit of his own graduate students; I could expect they might feel they owed me a favor in return some day. Accusing me of deliberately sabotaging one of his researcher's experiments was nearly unbearable; not least because the accusation was completely unfounded.
Now I was committed to spending the weekend before Thanksgiving, including my birthday, redoing some screw-up Master candidate's work so I could prove that I was innocent of malfeasance. What a farce. The worst part was that it was all statistics, which I hated. I'd seen math wizards who could make their slide rules fly, but I wasn't one of them.
I started after dinner, putting aside my own dissertation and research notes, and proceeded to cover my desk with neat stacks of paper. By the time I'd finished sorting, I'd remembered the experiment they described. It had been another deadly dull survey intended to measure attitudes across the student body; anybody with any excuse had contrived to be unavailable and Reynolds had started drafting the unwary -- like me -- to assist.
Reynolds' student, Alex, had claimed I had messed up my interviews and thrown off the entire study. More precisely, my data was skewed enough from the other interviewers' data that the uncertainty intervals became absurdly large. Removing my data reduced the population sufficiently that it was no longer possible to draw statistically significant inferences, even if the act of removing them didn't raise questions about the survey's methodology.
The survey was too simple to screw up. The interviewer showed the subject a pair of pictures, and recorded which was preferred. Then repeat about a hundred times. There were a lot of pictures, all carefully ordered and categorized so as to eliminate bias and allow conclusions to be based on the subject's demographic. It was deadly dull, but I
knew
I hadn't messed it up -- which meant the math claiming I did was wrong.
My problem was that by Saturday afternoon, it didn't look like the math was wrong after all. Sure, I'd done it five times and gotten three different answers, but I was beginning to think the accusation was correct -- or there was something subtly wrong with the experiment and nobody else had picked up on it. I changed tack and started looking for patterns in the data for my surveys.
I stumbled across it after dinner, and ended up awake well past midnight trying to confirm it. When I looked at my interviews in chronological order, I found the deviations were greatest with the first interviews of the day, and decreased until they became indistinguishable from the data collected by other interviewers. The other interesting quirk was that the deviations seemed to be generally in the same direction.
By Sunday afternoon, I had established a statistically significant trend existed; responses at the beginning of each day tended to converge, and responses at the end of each day tended to match the overall survey results. I also knew that I didn't know enough to take things any further. Since there was no way I was going to go to Reynolds and tell him that without knowing why, my obvious next step was to find Alex and talk to him.
I hurried through my own class Monday and let my students go a few minutes early so I could get across the quad before the end of the period. I'd never met him, but a glance at the schedule showed Alex was teaching a recitation section of Reynolds' Introduction to Psychology class; I figured it would be easy enough to intercept him at the end of the hour and introduce myself.
The students were already bolting from the classroom when I rounded the corner, so I let the mob pass before poking my head in the door. My first thought was that I'd missed Alex; the only person remaining in the room was a stunning blonde transferring some papers into a briefcase. I paused to admire the view for a moment, until it was clear she'd noticed me.
"Yes?" she prompted, obviously less taken with me than I with her. "Did you want something?"
"I was looking for Alex Sullivan; do you know where I can find him?"
The blonde barked a brief, unhappy laugh. "I'm Alexandra Sullivan -- what did you want?"
I walked a little further into the classroom. "I wanted to talk with you about your popular opinion survey." Her expression lightened, until I added, "My name is Lloyd Parker."
"You!" I think if she'd had something heavier in her hand than paper, she would have thrown it. "Do you know what kind of mess you've caused?"
Holding up both hands in self-defense, I admitted, "Yes; Dr. Reynolds pointed it out to me last week, rather forcefully!"
