Otago University Professor Richie Poulton, who co-authored the original research, said Dr Rogeberg's findings were based on simulations but his came from data.
Dr Rogeberg found that low socio-economic children's IQ's were temporarily boosted by schooling, but when they left school and became adults their IQ went back down to its former low baseline, Professor Poulton said.
If many cannabis users were formerly low socio-economic children, the coincidence could create the impression that cannabis was responsible for their drop in IQ as an adult.
However, Professor Poulton said that when looking at children from middle-class homes, the findings of a decline in IQ in adolescent-onset cannabis users remained unaltered.
"Many of the adolescent cannabis users in the [New Zealand] study were from slightly lower socio-economic homes, though not all.
"However, their IQs were not boosted by schooling, because their IQ scores are the same from the time they began school to adolescence," he said.
Professor Annette Dobson, director at the centre for longitudinal lifecourse research at the University of Queensland, said the approaches between the two papers differed.
"The original paper is led by authors specialising in psychology and psychiatry. They do not consider variables such as socioeconomic status, and their statistical analyses rely heavily on strong and untested assumptions.
"In contrast, Dr Rogeberg, an economist...takes socio-economic status into account, and has a more thorough approach to the statistical methods he applies. He is also clearer about the uncertainty of his results, using simulations to explore the potential effects of different assumptions," she said.
She said the original study did not have a large sample size which limited the extent it could take into account possible confounding factors, such as socio-economic status.
She could not say whether or not the original results were correct but said she agreed with Dr Rogeberg that the methods of analysis were inadequate.
The original study, which was published in the Proceedings of the National Academy of Sciences, examined individuals in the Dunedin Cohort study, which has followed 1037 people born in 1972-73 in Dunedin from birth.
About 5 per cent of the study group were considered cannabis-dependent, or used the drug more than once a week before age 18.
At age 38, all of the study participants were given a range of psychological tests to assess memory, processing speed, reasoning and visual processing. People who smoked cannabis persistently as teens scored significantly worse on most tests.
