I don't think the research is as clear cut as the article suggests. Firstly, the concept was created in the 60s and only became wide spread in the 80s. The study was conducted in the 1975.
This has two important implications:
- There were fewer people that were actually instructed in whole language and they skewed younger (and less practiced)
- The teaching profession had fewer years of as practitioners so methods resources were likely unrefined. Fewer books, instructional materials.
Also, there is always a bias to publish a scoop in acadamia, so unless there were multiple corroborating studies we should take it with a grain of salt.
Most importantly, I think that different kids learn differently. My son has been working on phonics for a long time and still struggles connecting sounds to words. In contrast, whole language approaches have been working better for him.