The data from published studies were used to build relationships between learning outcomes and air quality in classrooms. Psychological tests measuring cognitive abilities and skills, school tasks including mathematical and language-based tasks, ratings schemes and tests used to assess progress in learning including end-of-year grades and exam scores were considered to represent learning outcomes. Indoor air quality was characterized by concentrations of carbon dioxide (CO2). Short-term sick leave was included as well because it can influence learning. For psychological tests and school tasks, fractional changes in performance were regressed against the average concentrations of CO2 at which the changes were recorded; all reported data were used regardless of whether the change was statistically significant. For other learning outcomes and absence rates, the relationships created by original studies were used. The results predict that reducing CO2 concentration from 2,000 ppm to 1,000 ppm (equivalent to about 2.5 times higher outdoor air supply rate) would improve performance on psychological tests and school tasks on average by 12% (as regards the speed at which the teasks are performed) and by 3% (as regards errors made while performing the task. The performance on rating schemes will be improved by 1.3%. This change and will increase the number of pupils passing exams by 12%and is further estimated to result in about 6 out of 100 pupils improving their performance and to reduce absence by 0.5 day per student in a 200 days long school year.