The mathematical knowledge for teaching (MKT) measures have become widely used among researchers both within and outside the U.S. Despite the apparent success, the MKT measures and underlying framework have been subject to criticism. The multiple-choice format of the items has been criticized, and some critics have suggested that opening up the items might be an option. One way of opening up the items is to include commentary boxes that allow teachers to explain their thinking. This paper reports on a Norwegian study where commentary boxes were added to MKT items in order to investigate the connection between teachers' responses to the items and their written reflections. The results indicate that there is a mismatch between the answers given by the teachers on the MKT items and their written reflections. Teachers' written reflections do not always support their responses to the MKT items. [For the complete proceedings, see ED584829.]