I went to The Rochester Public Library (115 South Ave.) to check out the Thursday Thinkers lecture: Bob Spahn was to talk about The Earlier Springs, Later Falls in terms of whether birds are extending their migratory stay in the north owing to a warming climate. He rephrased the question: "do we have bird record data to show changes in arrival dates indicating warming climates?" and to that he said no.
He examined the data from several regional ornithological groups' records of arrival dates of birds in New York State but found no statistically significant difference in dates over 25 years of records. At first I wasn't satisfied with his analysis, but as the lecture continued, I realized it was the data that was the culprit. He mentioned that when you set out to analyze data, you need to carefully specify how to collect the data so it's relevant and useful.
The idea behind the data collection in New York State was to determine the number of each species of bird. As such, there were also notes of the first recorded sightings of each species. Additionally, the state is divided into 5-mile-by-5-mile regions for analysis. The criteria for observation seems specific, but even as Spahn noted: he recorded data for one region for several years until someone else took over and there was a huge disparity between their reports.
An alternative question to ask is, "is there a correlation between bird arrival dates and climate temperature?" The answer is "probably not". Spahn said that it's believed that bird migration is dictated by light levels. According to the WikiPedia article on bird migration, the purpose of migration is believed to do with the longer days in northern climates in the summer (i.e. more hours of daylight than the tropics) providing more hours to feed their young. Thus, even if there were climate change, it would likely not be shown in bird migration.