A clearly labeled opinion piece from some professor is hardly a "legitimate news source".
...or for that matter, a definitive statement on the nature of a discipline.
The reader will note that the Rosenberg op-ed cited above is principally focused on the macroeconomic predictive power of the dismal science. This clearly contrasts with the pedagogical research conducted by Fryer. Whether or not any failure on the macroeconomic side of economics necessarily impeaches the application of the discipline to pedagogical questions is a decision left to the reader.
In considering that question, I suggest that Rosenberg has misidentified the core essence of science. Rosenberg focuses on the predictive powers of science, but science is not prognostication. Science instead is a system of how to ask questions. In science a question, or hypothesis, is posed and it is challenged by research designed to answer that question. That's the essential element to science, the scientific method. While science may be used to make predictions, it is not the essential premise of that system. Here, Fryer used rigorous controls and tests to ensure that his study meet with the standards established by the scientific method.
In some disciplines, such as Newtonian physics as pointed out by Rosenberg, continued research utilizing the scientific method results in a body of knowledge capable of making accurate predictions. The application of the scientific method to other disciplines results in a weaker ability to make predictions, but this in no way makes the application of the scientific method less valid nor the research less scientific.
Furthermore, I remind the reader that the predictive power of early and preliminary research is limited. Early research into a topic cannot be expected to develop a hearty and indefeasible predictive model. Instead, early research into a topic is more frequently used to raise interesting questions and provide a body of work from which future robust predictive theories can be built. Newtonian physics is predictable in part because there is a healthy body of previous research. The reader may find that Fryer's research was a just such a preliminary study.
As a post script, readers may be interested to know that
I am not alone in believing that Rosenberg's view of science may not be in line with that of the population, general, intelligentsia, or academic.