Málfríður - 15.10.2008, Qupperneq 19

MÁLFRÍÐUR 1
of  clear  discouse  markers….to  infer  psychological 
or  metaphysical  meaning”  (OECD,  2007b,  p.  292). 
At Level 2 students can make use of low-level infer-
ences,  “construe  meaning  within  a  limited  part  of 
the text” and “follow logical and linguistic connec-
tions within a paragraph in order to locate or inter-
pret information” (OECD, 2007b, p. 293).
Of  the  eight  texts  in  this  assessment  two  are 
continuous  texts and  two are non-continuous  (one 
a table and the other a form). Four have a continu-
ous passage and non-continuous information in the 
form  of  a  table,  map,  diagram  or  chart.  Only  one 
text  could  be  described  as  literary.  The  rationale 
behind the diversity of PISA material is that:
PISA  measures  reading  literacy  in  terms  of  students’ 
ability to use written information in situations that they 
encounter in their lives. This goes beyond the traditional 
notion of decoding information and literal interpretation. 
Students are shown different kinds of text, and required 
to retrieve information, to interpret the text and to reflect 
on and evaluate what they read. (OECD, 2007a, p. 46)
PISA  designates  five  processes  in  reading  profi-
ciency (OECD, 2006): retrieving information; form-
ing  a  general  understanding  of  a  text;  developing 
an  interpretation;  reflecting  on  and  evaluating  a) 
the content, and b) the form of a text. Students are 
expected to be proficient to some degree in all these 
aspects of reading.
By  testing  these  students  in  English  students  I 
obtained  data  on  their  performance  in  Icelandic 
and, some eighteen months later, in English for the 
same  material. Attached  to  the  reading  test  was  a 
questionnaire  in  Icelandic  that  I  put  together  con-
taining  11  questions  and  a  self-assessment  also  in 
Icelandic  of  how  much  effort  students  had  put  in 
to  completing  the  reading  tasks.  Figure  2  shows 
the range of information I hoped to obtain from the 
questionnaire and its bearing on reading proficiency 
in English. 
Reading prociency in English
Reading prociency in Icelandic English language prociency
Metacognition
Effort
Self-assessment
Interest
Purpose
Topic knowledge
Figure 2. Factors affecting reading in English
Written permission was obtained from the schools 
involved,  parents  or  guardians  of  students,  and 
students  themselves.  Anonymity  of  participating 
institutions  and  students  was  assured.  Students 
received  no  remuneration  for  taking  part  in  the 
study. Test scores were analysed, compared and cor-
related with the questionnaire.
Results
Table 1 below shows student scores for the 28 ques-
tions of  the  reading  tests  in  Icelandic and English, 
for  the  whole  group  of  students  and  separately 
for  the  weaker  and  stronger  readers  in  Icelandic. 
Students who classed as better readers in Icelandic 
in  PISA  2006  scored  higher  on  the  test  as  a  whole 
(M = 7.65, SE = 0.11) than they did in English (M = 
6.51, SE = 0.27,  t(31) =  -4.39, p <  .05,  r =  .62, diffe-
rence  between  means  =  1.14).  Performing  well  in 
Icelandic correlated strongly with performing well 
in English. 
The breadth of difference among the weaker read-
ers in Icelandic was noticeably greater than among 
the stronger readers in Icelandic. For students who 
performed less well in Icelandic in PISA 2006 there 
was  no  significant  difference  between  scores  in 
Icelandic  on  the  whole  test  (i.e.not  only  the  ques-
tions students answered) (M = 4.22, SE = 0.17) and 
English (M = 3.90, SE = 0.46, t(24) = -.67, p = .51 (2-
tailed), r = .13, difference between means = .31). 
It would seem that these students’ reading skills 
in  the  first  language  transfer  at  almost  the  same 
level to the second language.
Figure 3 on the next page shows the difference in 
individual scores in Icelandic and English for each 
student at the higher and lower reading ability level 
in Icelandic separately.
The broad variance  in difference between scores 
in Icelandic and English among the weaker readers 
in  Icelandic  suggests  less  reliability  in  their  per-
formance, which may  lead  to  students performing 
better  or  worse  on  another  occasion.  These  charts 
show clearly  that many students performed better 
Mean scores measured over the whole test
Weaker
readers in
Icelandic
Stronger
readers in
Icelandic
All students
Reading in Icelandic 4.22 7.65 6.14
Reading in English 3.90 6.51 5.37
Table 1. Scores measured over the whole test