I've seen that study, too. As the article mentions, it's possible the discrepancy is a result of demographics--i.e., people who can't afford to hire movers are more likely to rent a uhaul, which means the uhaul data will be weighted more heavily in favor of states that people with less money are likely to move to.

Unfortunately, all this data is proprietary, either owned by uhaul, or united van lines. I'm not aware of any independent researchers gathering broad, normalized market data on this. I'd be very interested to see it, if it's out there.