In a major advance over the inherently flawed effort to use test score growth estimates to measure teaching and learning, Big Data is pioneering the next step in identifying the characteristics of effective teaching.
EdTweak’s Harvard Teacher Team Links Teacher Traits to Value-Added describes a groundbreaking research design using a data set including three years of Google searches.
It reports that “Harvard Professor Sage Petty and his colleagues were able to determine that teachers with higher value-added scores were 0.0408% more likely to prefer Mary Ann to Ginger, 0.0783% more likely to purchase their firearms at discounted prices, and 0.0281% more likely to be able to distinguish a Mallard from a Fulvous Whistling-Duck.”
Petty documents other “really amazing the sorts of associations one can tease out with a large enough data set.” He documents correlations between value-added scores and “purchases of laundry detergent (powered-detergent teachers have higher scores) and searches combining ROTFL and IMHO (lower scores)."
Petty is surprised by the finding, “teachers at all value-added levels had an equal likelihood of wanting to slap me and my colleagues upside the head with a trout.”-JT(@drjohnthompson) Image via.