comprehension strategies through RT. They did not even learn about the RT approach in
this course. The teaching methods used in this condition included, for example, teachers’ and
students’ lectures, self-study, and group discussions about case examples.
Measures
Reading-related measures.
Two measures were administered in the study at pre- and posttest.
In the absence of a standardized reading test for adults, the Lesegeschwindigkeits- und -
versta¨ndnistest fu¨r die Klassen 6–12 [Reading speed test and reading comprehension test for
grades 6–12] (LGVT 6–12 see Schneider, Schlagmu¨ller, & Ennemoser, 2007) was used. The
LGVT 6–12 is a single choice test, originally developed for sixth- to twelfth-grade students.
Participants had to read an expository text of about three and a half pages in length. Some
sentences in the text had blanks with three words given in brackets. In order to select and
underline the word that best fit the content of the text, it was necessary for the test taker to
read the text thoroughly and comprehensively rather than only superficially (e.g., These
[heads, spots, horns] are covered by skin and end in tufts.) Because the LGVT 6–12 does
not provide A and B versions of the test, the same text was used at both measurement points.
The length of the text was deliberately designed so that nobody would be able to read it
completely within the allotted processing time. After finishing the test, participants were
asked to draw a vertical line behind the last word they read. Reading speed could thus be
measured by counting all of the words that had been read, and the number of words per
minute could be calculated afterwards. Reading comprehension was measured by adding up
all of the correctly underlined words. To calculate the raw points, two points were given for
every word that was underlined correctly. One point was deducted for any word underlined
incorrectly. If no word was selected out of the three given words, the test taker was given
zero points for that item. For both measures, the raw points were converted into standard
scores, with
M
¼
50 and
SD
¼
10. The reference population consisted of twelfth-grade stu-
dents attending academic high schools. Standard scores between 40 and 60 define a reader as
comparatively average. Scores greater than 60 indicate above-average performances,
whereas scores of less than 40 indicate below-average performances. Reliability
(Cronbach’s Alpha) in the current study sample varied from .86 to .93 for both reading
speed and reading comprehension at both time points.
Social validity.
At posttest, RT students were asked whether they would use RT to improve the
reading skills of their primary school students in their own future lessons. They indicated
their response to this item on a 3-point Likert scale (0
¼
no, 1
¼
I do not know yet, 2
¼
yes).
We used this item to measure the social validity.
Treatment integrity.
To ensure the integrity of the treatment (Gresham, 1989) of the RT imple-
mentation, fidelity measurements were established through written self-reports provided by
the trainer. Following the instructions from the training manual, the trainer had an overview
of all single content points for each training lesson. In every lesson, the trainer had to rate
whether each content point had been realized as intended. There was a 3-point Likert
scale (0
¼
not realized, 1
¼
Do'stlaringiz bilan baham: