Creative Education
2011. Vol.2, No.4, 381-387
Copyright © 2011 SciRes. DOI:10.4236/ce.2011.24054
Impact of Asking Support Questions on Grades 4 and 7
Students Reading Comprehension
Pilve Kängsepp
Department of General Education, Faculty of Social Science and Education, Tartu University,
Tartu, Estonia.
Received August 14th, 2011; revised September 18th, 2011; accepted September 26th, 2011.
This study investigated the effect of support questions and their timing on Grades 4 and 7 students’ reading
comprehension. The support questions were posed either during or after reading texts. A comparative analysis of
students’ scores in answering of control questions presented at the end of the reading assignment revealed that
asking inferential support question has positive effect on the text comprehension only in certain conditions. The
effect of inferential support questions is rather negative if the text to be read is difficult and support questions are
posed after reading the text sections.
Keywords: Reading, Reading Comprehension, Question Timing, Questions
Introduction
Reading comprehension has been a part of classrooms as
long as there have been schools, texts, students, who desire to
read them, and teachers wanting to both promote and assess
their understanding (Pearson, 2009). Students are working with
texts for about 60% of the lesson time and most of their home-
work consists of studying texts (Johnsen, 1993). The compre-
hension of a text can range from the most superficial to deep
understanding but the school learning requires a rather deep
understanding of study texts (Kintsch, 1994). The majority of
students comprehend study texts without difficulty but there are
always some children for whom reading comprehension is a
particularly difficult task (Spooner, Gathercole, & Baddeley,
2006; Megherbi, Seigneuric, & Ehrlich, 2006). Furthermore,
Oakhill and Yuill (1996) pointed out that a reason for many
students’ learning difficulties is poor understanding of written
texts.
A number of factors interfering with the development of stu-
dents’ text comprehension skill have been uncovered by re-
searchers. These factors range from efficiency of decoding and
catching meaning at the level of single words through syntactic
development to inference making and integration of the ideas in
the text as whole (Oakhill & Cain, 2003). Yet, Nation and An-
gell (2006) pointed out that a greater part of difficulties in
reading comprehension remain hidden and unnoticed by teach-
ers.
Good text comprehension by students means that they inte-
grate the text information with their previous knowledge and
acquire new information. One possibility to support reading
comprehension process might be answering questions.
The purpose of this study is to investigate the impact of sup-
port questions on text comprehension depending on students’
grade, timing of support questions and text complexity.
Text Comprehension
In the theory of text comprehension of Van Dijk and Kintsch
(1983; Kintsch, 1994) the process of reading with understand-
ing is conceived as construction of semantic representations of
the text (text-base model) and also as creating of more personal
mental representations integrated with the reader’s prior knowl-
edge (situation model). The text-base level memory-based rep-
resentations of the words and ideas are constructions as they
appear in a text. The words and phrases themselves are encoded
as linguistic relations between them. The text-base level of
representation would include a propositional description of the
explicit text (Kintsch & Van Dijk, 1978), which means the
text-base level understanding provides a shallow representation.
In the situation model the information extracted from the text is
elaborated from the prior personal knowledge and is integrated
with it (Kintsch, 1994). This is why Van Dijk and Kintsch have
distinguished text-base model from the situation model. When
we read the text we do not perceive the individual sentences
conveying isolated pieces of information. Rather, we perceive
them as being interconnected, as forming a coherent whole. The
perception of coherence is the result of extensive inferential
processes, some of which are automatic while others are inten-
tional and strategic. A representation results from extensive
inferential processes that take place during reading (Van den
Broek, Rohleder, & Narvaez, 1994). In general, inference is a
cognitive process used to construct meaning (Davoudi, 2005).
Therefore, it is widely accepted by researchers that the ability
to make inferences is necessary for reading comprehension (e.g.,
Oakhill & Cain, 2003; Davoudi, 2005). Self-evidently, if the
reader does not understand the text-base, the situation model
will not be adequately constructed. In this case, the reader could
be able to answer factual questions, but he or she is not capable
to make inferences. Or, expressed in another way, students have
a good understanding of textual material, when they have suc-
cessfully integrated information from the text with their prior
knowledge, and thus are able to learn the new material (Koz-
minsky & Kozminsky, 2001).
Paris, Cross and Lipson (1984) showed that some students in
elementary education seemed to have difficulty in understand-
ing the relationship between evidences and conclusions and this
may make it difficult to see how words can provide clues to
meaning within text. Long, Oppy and Seely (1994) argued that
if readers are unable to generate inferences that connect explicit
information in a text to relevant prior knowledge, they feel as

P. KÄNGSEPP 382
though they do not comprehend the text and have difficulty
remembering it. They are often able to integrate information at
the level of single sentences but are unable to produce a coher-
ent integrated model of the text as whole (Mannes & Kintsch,
1987). This is because poor readers have difficulties with mak-
ing inferences and these difficulties are likely a cause of their
understanding of texts at a low level of generalization (Cain &
Oakhill, 1999; Perfetty, Landi, & Oakhill, 2005).
A deeper comprehension is achieved when the reader con-
structs causes and motives that explain why an event or action
occurred, by inferring the global message or point of the text,
and by relating the state of affairs or situation described in the
text to the state of affairs in the world which is located in the
background knowledge structure of the reader of the text. A
deeper comprehension means that the reader constructs more
levels of representations and makes more inferences than it is
possible reading at the text-base level only (Davoudi, 2005).
Consequently, if the aim is improving the reader’s text com-
prehension, it is necessary to challenge him or her to construct
representations and make inferences of this text. Sundbye (1987)
found that asking inferential questions about relationships be-
tween characters in a text, their goals and motivations for action,
enhanced children’s comprehension as effectively as modifying
the text to be more explicit. Inferential questions call for a more
thorough and systematic processing of a text and prompt the
integration of text-based concepts with a broader background
knowledge, which may enhance children’s understanding of a
text (McGee & Johnson, 2003). This means that asking of in-
ferential questions may be a helpful tool in supporting students’
reading with understanding.
Using Questioning for Increasing Text
Comprehension
Many studies refer to an opportunity of increasing children’s
reading with understanding by having them to answer questions
pertaining to the text to be learned (Anderson & Biddle, 1975;
Pressley et al., 1989; Cerdan et al., 2009). According to Olson,
Duffy, and Mack (1985) answering of these questions can be
considered as an indicator of understanding the text to be read,
but posing of questions can also serve as an advance organizers
of meaning (Ausubel, 1968) in further reading of the text.
Questions support the construction of a causal network repre-
sentation, resulting in better understanding of a text as a whole.
This suggests that asking the reader some questions while read-
ing a text might be an effective tool for guiding his or her un-
derstanding of the text. For example, Van den Broek et al.
(2001) showed that posing supportive questions during reading
helps understanding of the text. Answering questions pertaining
to the text to be read enhances learning from it (Pressley et al.,
1989) because the questions challenge the reader to find rela-
tions between queried and answered information and focus the
reader’s attention on moments necessary for comprehending the
text (Olson, Duffy, & Mack, 1985). However, the positive ef-
fect of asking support questions does not appear uncondition-
ally. It is the most effective if it really helps a reader to focus on
finding relations between perceived and resulting information
that he or she is expected to have as a prove of comprehending
the read text. In certain cases asking of support questions may
cause cognitive overload of student short term memory and be
ineffective (Kirschner, 2002). One among many factors that
would impact the readers’ information processing capabilities is
timing of support questions (Van den Broek et al., 2001). For
example, if finding answers to support questions causes cogni-
tive overload during reading, it barely supports understanding.
Therefore, the timing of questions for supporting text com-
prehension in reading has caught attention of many researchers
(see Anderson & Biddle, 1975; Pressley et al., 1989; Van den
Broek et al., 2001). Questions can be asked before, during and
after reading the text. Questioning after reading the entire text
generally increases text comprehension in adults (Anderson &
Biddle, 1975) and helps only if questions lead to reprocessing
of relevant information after the reader fails to answer an after
text question. Pressley et al. (1989) showed that younger stu-
dents are unable to look back for finding answers to after text
questions that they could not answer directly. This failure is
common to elementary school children and poor readers even
in upper elementary school (Pressley et al., 1989).
Van den Broek et al. (2001) demonstrated that questions that
were presented during reading helped the older students to
memorize information targeted by the questions. The same
effect was not documented with younger students. The explana-
tion is in the light of the notion of cognitive overload that ques-
tioning during reading calls for a simultaneous processing of
general information drawn from the text and of specific infor-
mation targeted by the questions. Due to their poor reading
skills, younger readers fail to implement these two tasks simul-
taneously, and presenting of inferential questions asked during
reading rather interferes with their text comprehension than
supports it. This means that not all ways of questioning are
helpful for reading comprehension, a conclusion that is con-
firmed by conflicting results of former research (see e.g. Ander-
son & Biddle, 1975; Pressley et al., 1989; Van den Broek et al.,
2001; Cerdan et al., 2009).
The specific aim of this study is to investigate the impact of
support questions on academic text comprehension in Estonian
language depending on students’ grade level and timing of
questions in classroom. The research questions of this study
were the following. First, how timing of inferential support
questions influences reading comprehension depending on the
age of readers; second, how does the impact of support inferen-
tial questions depend on text complexity in Estonian condi-
tionsns.
Methods
Participants and Procedure
This study was carried out in Estonian comprehensive school
consisting of students from schools willing to co-operate. The
study was conducted in the beginning of the school year, with
students who had completed Grade 3 and 6, and started a new
school level (Grade 4 and 7, respectively). 274 students par-
ticipated in the first phase and 294 in the second phase of this
study (Table 1). There were 131 boys and 111 girls among the
Grade 4 students, and 157 boys and 169 girls in Grades 7. The
students’ average age in Grade 4 was 10.5 years and in the
Grade 7 13.4 years. All of them were native Estonian language
speakers.
The impact of asking support questions on student reading
comprehension was tested on the basis of two texts with differ-
ent readability indices in both grades. In the phase one, carried
out in 2007/08, a rather easy readable text, and in phase two, in
2008/09 school year, a more difficult text was used for testing.
In the both research phases participants were divided (at ran-
dom basis) into three groups at each grade level—experimental
group 1 and experimental group 2 and a control group.

P. KÄNGSEPP 383
Table 1.
The distributions of students by g ra d es and groups.
Phase I Phase II
Grade 4 Grade 7Grade 4 Grade 7
In All
Experimental
group I 37 52 42 53 184
Experimental
group II 36 57 43 58 194
Control group 42 50 42 56 190
In all 115 159 127 167 568
The experimental and control groups received their text book-
lets and after the reading the instruction participants progressed
through the booklets at their own pace according to the instruc-
tion. The students in the experimental groups were asked to
read the text in two different conditions, and to answer control
questions for testing their understanding afterwards. In the ex-
perimental group 1 the comprehension of the text was sup-
ported by asking inferential support questions after reading the
text just before answering the control questions. In the experi-
mental group 2 the support questions were posed during read-
ing the text after reading through its sections. The students in
the control group answered control questions after reading the
text.
Characterization of Text and Measurements
A story titled “Catching words” written by H. Veigel from
third grade textbook was adapted for appropriate for phase one
of this study. This is an expository text about invention of the
recorded messages. A readability index of the text1 was 13.0
(easy readable text) that is within powers of Grade 3 students
according to the established criteria of optimality for the Esto-
nian readability index (Mikk, 2000). The text is 27 sentences
long and its average length of sentences is 10 words.
For the phase two a story about insect life on the earth, writ-
ten by U. Tartes, also from third grade textbook but with the
readability index 15.0 (hard readable text), was selected. This
text is 38 sentences long and its average length of sentences is
12 words.
The readability indices of potential texts in this study were
determined using Mikk’s (2000) readability formula. The text
complicacy index includes independent sentence length in letter
spaces and abstractness of repeating nouns in the text passage
according to three scale. Total frequency of text’s words ac-
cording to the list of known words divided by the number of
text’s words (Mikk, 2000).
All materials necessary for investigating reading comprehen-
sion of experimental and control groups were delivered to stu-
dents in special booklets. Three different versions of booklets
were constructed for presenting support questions to experi-
mental groups and for presenting control questions in each
phase of the study. The control questions were located overleaf
in regard of texts to be read in order to prevent consulting the
text when answering the questions.
In the both phases of the study five support questions and the
same number of control questions were used. The decision for
using five inferential support and control questions was based
on the analysis of texts by a triad of primary school teachers
who came to a consensual agreement that the both texts contain
five paragraphs that could considered as logical entities. The
preliminary sets of support and control questions were also com-
piled by the same team of teachers. To this end they first re-
corded all possible questions pertaining to these texts. Then
selected by teachers on consensual basis the most suitable ver-
sions for support and control questions. The consensus was of
at least 92%.
Answers to support and control questions were scored on bi-
nary scale by assigning each correctly answered question one
point and incorrectly answered questions zero point. Keeping in
mind that the answers to inferential support questions might
have interpretative character; a scoring guide was compiled
with prototypes of correct answers. In total a student could obtain
5 points maximum for the correctly answered support questions
as well as for answering the control questions. Next, the ob-
tained data, including quantified demographic data of students
were tabulated and processed using methods of descriptive and
inferential statistics.
Results
Comparison of Mean Scores in Answering Control
Questions
The mean scores of the experimental and the control groups
in answering the control questions in Grades 4 and 7 are given
in Table 2 for the phase one and in Table 3 for the phase two of
the study. The comparison of the mean scores of the control
questions in phase one shows that in the experimental groups of
Grade 4 there was no statistical significant difference found
between the groups. Yet, in the experimental groups of Grade 7
the exposure to the support questions caused a statistically sig-
nificant increase in reading comprehension in comparison with
the control group. In the experimental condition one when the
support questions were presented after reading the text, the
average scores of answering control questions was 2.9 but in
the experimental condition two when the support questions
were posed during reading, the mean score in answering the
ontrol questions was 2.8 in the control group was 2.2 (Table 2).
In the phase two of the study the mean scores of answering
the control questions in the control groups (2.0 and 2.5) were
higher than in the experimental groups at the both grade levels.
The differences were statistically significant. It means that an-
swering of inferential support questions, independently on tim-
ing, aggravated comprehension of the text with higher readabil-
ity index (Table 3).
Comparison of Mean Scores of Experimental Groups
in Answering of Inferential Support Questions
The mean scores in answering the support questions by ex-
perimental groups in phases one and two of the study are pre-
sented in Table 4. As can be seen from the table, the average
scores in answering the support questions were statistically
significantly lower in the phase two than in the phase one for
the both experimental conditions. The differences in scores
were the biggest for the experimental condition one (0.9 for
Grade 4 and 1.4 for Grade 7) when the support questions were
posed after reading the text. The significantly lower scores in
answering the support questions in the phase two (when a text
with a higher readability index was used) confirm that the
students had more difficulties in answering support questions
n this case than in the phase one.
1Readability index is an index of text complicacy while higher values o
f
readability index show that the text is less readable (Mikk, 2000). i

P. KÄNGSEPP 384
Table 2.
Comparison of mean scores in answering control questions by experimental an d c o n t rol groups in phase one (easy readable text).
Control Group Experimental Group I (Support Questions After Reading) Experimental Group II (Support Questions During Reading)
Grade
Mean (SD*) Mean (SD) Diff** p Mean (SD) Diff p
4 1.8 (1.3) 2.1 (1.1) 0.3 >0.05 1.4 (1.2) −0.4 > 0.05
7 2.2 (1.3) 2.9 (1.3) 0.7 <0.02 2.8 (1.4) 0.6 < 0.02
*SD—Standard deviation; **Diff—Difference with the mean score of the control group.
Table 3.
Comparison of mean scores in answering control questions by experimental an d c o n t rol groups in phase two (hard readable text).
Control Group Experimental Group I (Support Questions After Reading) Experimental Group II (Support Questions During Reading)
Grade
Mean (SD*) Mean (SD) Diff** p Mean (SD) Diff p
4 2.0 (1.0) 1.6 (0.8) −0.4 <0.03 1.4 (1.0) −0.6 <0.00
7 2.5 (1.1) 2.0 (1.2) −0.5 <0.04 2.2 (1.0) −0.3 <0.06
*SD—Standard deviation; **Diff—Difference with the mean score of the control group.
Table 4.
Comparison of differences in mean scores of experimental groups in answering supp o r t questions in the phases one and two.
Experimental Condition I (Support Questions after Reading) Experimental Condition II (Support Questions during Reading)
Grade
Mean Phase I Mean Phase II Diff p Mean Phase I Mean Phase II Diff p
4 4.0 3.1 0.9 <0.00 3.9 3.4 0.5 <0.03
7 4.4 3.0 1.4 <0.04 4.4 3.4 0.4 <0.05
The comparison of mean scores in answering support ques-
tions at different timing conditions revealed no dependence on
timing of support questions in the phase one of the study (Table
5). The difference in mean scores does not surpass 0.1 points.
Instead, in the phase two where a more difficult text with higher
readability index was used, the students tended to produce bet-
ter result in answering support questions when they were pre-
sented during reading the text. In Grade 4 the difference was
0.3 points (not statistically significant) and in Grade 7 the dif-
ference was 0.4 point that is statistically significant (p < 0.02).
Correlation Answe ring C ontrol and Support
Questions
The correlations between average scores of answering sup-
port questions and control questions for the phase one in Grade
4 and 7 were r = 0.34 and r = 0.47 (when support questions
asked after reading) and r = 0.48 and r = 0.60 (when support
questions asked during reading), see Table 6.
The relationships between average scores of answers given to
support and control questions for the phase two of the study in
Grade 4 and 7 were r = 0.33 and r = 0.32 (when questions asked
after reading), and r = 0.47 and r = 0.16 (when questions asked
during reading). All values are statistically significant except
0.16.
Discussion
Our comparative analysis of students’ scores in answering of
control questions taken as indicators of comprehension of texts
to be read reveals that the impact of asking inferential support
questions depends on the reading proficiency (correlating with
the age of students in general) and on the timing (i.e. asked
questions during or after reading) of these questions. The effect
was stronger in the phase one when an easy readable text (op-
timal for Grade 3 level students) was used. However, even in
this condition, the impact of inferential support questions de-
pends on the students’ grade level as well as on the timing of
support questions. Asking inferential support questions im-
proved students’ comprehension of the text in Grade 7 inde-
pendently on the timing of support questions. Instead, the ex-
posure of Grade 4 students to inferential support question did
not show significant differences in reading comprehension
between control and experimental groups. However, it can be
said with certain reservation that inferential support questions
posed after reading had a rather positive effect on text compre-
hension.
When a text with a higher readability index was used in the
phase two of the study, asking of support questions did not
have any positive effect on the text comprehension at either
grade levels. Instead, the mean scores in answering control
questions were significantly lower at both grade levels. Fur-
thermore, the higher scores of the control groups in answering
control question in comparison with the experimental groups at
both grade levels points to the fact that answering of inferential
questions rather interfered with remembering of essential in-
formation. It occasionally caused a kind of cognitive overload
in students or distracted them from remembering factual infor-
mation. The explanation might be that answering of support
questions during reading may overload the young readers’
working memory because they have to deal simultaneously
with their normal reading processes and with answering the
questions. Questions posed after reading also tax young read-

P. KÄNGSEPP 385
Table 5.
Comparison of mean scores o f experimental groups in answering support que s t io n s w i th in the phases one and two.
Phase I Phase II
Grade
Mean Exp C I* Mean Exp C II** Diff p Mean Exp C I Mean Exp C II Diff p
4 4.0 3.9 0.1 >0.05 3.1 3.4 −0.3 >0.05
7 4.4 4.4 0.0 >0.05 3.4 3.8 −0.4 <0.02
*Exp C I stands for “Experimental condition I (support questions presented after reading)”; **Exp C II stands for “Experimental condition II (support questions presented
during reading)”.
Table 6.
Correlations between scores of answering inferential support and con-
trol questions by grade levels, research phases, and research condi-
tions.
Phase I Phase II
Grade
Exp C I* Exp C II** Exp C I Exp C II
4 0.34 0.48 0.33 0.47
7 0.47 0.60 0.32 0.16***
*Exp C I stands for “Experimental condition I (support questions presented after
reading)”; **Exp C II stands for “Experimental condition II (support questions
presented during reading)”; ***All values in Table 6, except 0.16, are statistically
significant at p < 0.05.
ers’ ability to construct a coherent representation, but such ef-
fect is weaker than that of questioning during reading (Van den
Broek et al., 2001).
Though the only clearly documented factor that might elimi-
nate the positive impact of support questions on reading com-
prehension in the phase two was using of a text that was more
difficult to read and understand, other factors like students’
familiarity with text topics or dependence on the nature of sup-
port and control questions cannot be ruled out either. For ex-
ample, the topic of the text used in phase one could be more
familiar to the students than that used in phase two. Van den
Broek et al. (2001) also pointed out that questioning might in-
terfere even with skilled readers’ comprehension if the reading
task is highly demanding, for example when the content or str-
ucture of the text is complicated or unfamiliar to reader.
It also cannot be ignored that in the phase two not only the
readability index of the used text was higher, but also this text
was longer (38 sentences) than that used in phase one (27 sen-
tences). The fact that this text called for higher reading and
memorizing skills was also corroborated by a lower mean score
in answering control questions by the total research sample in
comparison with phase one. The same conclusion can be made
from the fact that negative and significant correlations of scores
in answering support and control questions with the phase num-
ber of the study were uncovered. Also, when calculated sepa-
rately for grade levels, the mean score for Grade 7 in phase one
was significantly higher than in phase two. However, the dif-
ference was less than half point. A comparison of mean scores
in answering of support questions between two phases shows
that these scores are significantly higher in the both experimen-
tal conditions for phase one. At the same time only one relevant
difference in the mean scores in answering support questions
caused by their timing was uncovered at Grade 7 level for phase
two i.e. the students had higher scores in answering the support
questions when they were posed during reading although the
difference was also here less than a half point.
Quite informative are correlations found between scores in
answering support and control questions. The fact that the cor-
relation was significantly higher for the phase one than for the
phase two confirms that there was less impact of answering
support questions on answering control questions. This conclu-
sion is also supported by differences of correlations that are
calculated separately for Grades 4 and 7 for the both phases of
the study. In the phase one when the text was more within
powers of Grade 4, the correlations in answering these two types
of questions for Grades 4 and 7 were almost equal. In the phase
two the correlation was still statistically significant for the Grades
4 and 7. A more detailed analysis of the last mentioned correla-
tion revealed that in the experimental condition one this corre-
lation was for Grade 7 but was insignificant for the condition
two (when the support questions were asked during reading).
A low correlation between scores of answering support and
control questions in phase two at Grade 7 level in comparison
with phase one points to the fact that there is practically no
impact of support questions on the comprehension of the entire
text when the support questions are presented during reading
the text. Furthermore, the difference between the means corre-
lation quotients is statistically significant. A possible explana-
tion for this situation is that the text used in phase two was not
only beyond powers of Grade 4 students but also of Grade 7
students.
However teachers should be careful in using support ques-
tions until students reach an adequate level of reading profi-
ciency (see also Seretny & Dean, 1986; Tal, Siegel & Maraun,
1994; Van den Broek et al., 2001). Since teachers mostly rely
on short-answer or completion test items in their everyday work
with students than use questions that require answering with
phrases or complete sentences it is possible that they do not
even notice that the text as a whole is not understood by the
students. Thus, the reading comprehension difficulties remain
unnoticed as Nation and Angell (2006) have pointed out.
Yet, the teachers must teach students to learn textual infor-
mation by relating it to something they already know. Some
students do not understand how to construct the answer from
prior knowledge and textual content. Thus, asking inferential
support questions may become a powerful tool in promoting
children’ text-related activities. Answering to inferential sup-
port questions imposes a more thorough and systematic proc-
essing of text and prompt the integration of text-base concepts
with background knowledge, which may enhance children’s
comprehension of a text. Poor readers may need a systematic
training in such skills (McGee & Johnson, 2003; Davoudi, 2005).
These conclusions should be taken with some precaution be-
cause of some limitations of this study. For making them more
reliable new studies are needed that take the characteristics of
the texts (in terms of familiarity to students and length) with
different readability indices better under control. Also, experi-

P. KÄNGSEPP 386
mentation with students in other elementary grades could be
beneficial for more reliable conclusions on the effect of an-
swering inferential support questions on promoting reading skills.
The limitations in tests (no normative test only teachers choices
of questions, no memory test, no vocabulary test, no fluency
test); possible floor effect and ceiling effect of scores, and the
limitations of the analysis that no predictive relationships can
not be said. So the results remain in descriptive level but they
give some important aspects to study questioning in compre-
hension in the future.
Conclusion
In summary asking inferential support questions has positive
effect on the text comprehension only in certain conditions. The
impact of support questions depends on the students’ grade
level as well as on the timing of support questions. Our findings
also suggest that asking of support questions can be used for
directing the attention to specific information and to prompt to
specific connection. The outcomes of this study have practical
implications for elementary grade teaching of reading compre-
hension skills by using support questions. Especially, these
outcomes might be helpful in providing texts (study aids) used
for exercising reading comprehension with inferential support
questions. When preparing these assignments, the combined
impact of the both factors—readability of the underlying text
and timing of inferential support questions—must be taken into
account. When using an easy readable text answering of infer-
ential support questions promotes text comprehension rather
independently of their timing. In the case of texts that are
harder to read but still within the power of students, inferential
questions that are posed after reading might more effective for
promoting reading comprehension. When the text is clearly
beyond the capabilities of students, answering of inferential
support questions generally has no positive effect on text com-
prehension and corresponding reading assignments can barely
promote students’ skills of reading with understanding.
Acknowledgements
The author thanks Edgar Krull for his comments and advice
in preparation of the article.
References
Anderson, R., & Biddle, W. (1975). On asking people questions about
what they are reading. In G. H. Bower (Ed.), The Psychology of Learn-
in g and Motivation, (pp. 90-132). New York: Academic Press.
doi:10.1016/S0079-7421(08)60269-8
Ausubel, D. (1968). Educational psychology: A cognitive view. New
York: Holt, Rinehart & Winston.
Cain, K., & Oakhill, J. V. (1999). Inference making ability and its rela-
tion to comprehension failure in young children. Reading & Writing:
An Interdisciplinary J ournal, 11, 489-503.
Cerdan, R., Vidal-Abarca, E., Martinez, T., Gilabert, R., & Gil, L. (2009).
Impact of question-answering task on search processes and reading
comprehension. Learning and Instruction, 19, 13-27.
doi:10.1016/j.learninstruc.2007.12.003
Davoudi, M. (2005). Inference generation skill and text comprehension.
The Reading Matrix, 5, 106-126.
Johnsen, E. B. (1993). Textbooks in the kaleidoscope. A critical survey
of literature and research on educational texts (pp. 165-166). Oslo:
Scandinavian University Press.
Kintsch, W. (1994). Comprehension, memory and learning. American
Psychologist, 4, 294-303.
doi:10.1037/0003-066X.49.4.294
Kintsch, W., & Van Dijk, T. A. (1978). Toward a theory of automatic
information processing in reading. Cognitive Psychology, 6, 294-323.
Kirschner, P. A. (2002). Cognitive load theory: Implications of cogni-
tive load theory on the design of learning. Learning and Instruction,
12, 1-10.
doi:10.1016/S0959-4752(01)00014-7
Kozminsky, E., & Kozminsky, L. (2001). How do general knowledge
and reading strategies ability relate to reading comprehension of high
school students at different educational levels. Journal of Research
in Reading, 24, 187-204. doi:10.1111/1467-9817.00141
Long, D. L., Oppy, B. J., & Seely, M. R. (1994). Individual differences
in the time course of inferential processing. Journal of Experimental
Psychology: Learning Memory and Cognition, 20, 1456-1470.
doi:10.1037/0278-7393.20.6.1456
Mannes, S., & Kintsch, W. (1987). Knowledge organisation and text
organisation. Cognition and Instruction, 4, 91-115.
doi:10.1207/s1532690xci0402_2
Megherbi, H., Seigneuric, A., & Ehrlich, M. F. (2006). Reading com-
prehension in French 1st and 2nd grade children: Contribution of
decoding and language comprehension. European Journal of Psy-
chology of Education, 21, 135-147. doi:10.1007/BF03173573
Mikk, J. (2000). Textbook: Research and writing. Frankfurt am Main:
Peter Lang.
Nation, K., & Angell, P. (2006). Learning to read and learning to com-
prehend. London Review of Education, 4, 77-87.
doi:10.1080/13603110600574538
Oakhill, J., & Yuill, N. (1996). Higher order factors in comprehension
disability: Processes and remediation. In J. C. Cornoldi and J. Oakhill
(Eds.), Reading Comprehension Difficulties. Processes and Interven-
tion (pp. 69-92). Mahwah, New Jersey: Erlbaum.
Oakhill, J. V., & Cain, K. (2003). The development of comprehension
skills. In P. Bryant (Ed.), Handbook of Children’s Literacy (pp. 155-
180). The Netherlands: Kluwer Academic Publisher.
Olson, G. M., Duffy, S. A., & Mack, R. L. (1985). Questions-asking as
a component of text comprehension. In A. C. Graesser and J. B. Black
(Eds.), The Psychology of Questions (pp. 219-226). Mahwah, New
Jersey: Erlbaum.
Paris, S. G., Cross, D. R., & Lipson, M. Y. (1984). Informed strategies
for learning: A program to improve children’s awareness and com-
prehension. Journ al o f Educational Psychology , 76, 1239-1252.
doi:10.1037/0022-0663.76.6.1239
Pearson, P. D. (2009). The roots of reading comprehension instruction.
In S. E. Israel and G. G. Duffy (Eds.), Handbook of Research on
Reading Comprehension (pp. 3-31). New York: Routledge.
Perfetty, C. A., Landi, N., & Oakhill, J. (2005). The acquisition of read-
ing comprehension skill. In M. J. Snowling and C. Hulme (Eds.), The
Science of Reading: A Handbook (pp. 227-253). Oxford, UK: Black-
well. doi:10.1002/9780470757642.ch13
Pressley, M., Johnson, C. J., Symons, S., McGoldrick, J. A., & Kurita, J.
A. (1989). Strategies that improve children’s memory and compre-
hension of text. The E lementary School Journal, 90, 3-32.
doi:10.1086/461599
Seretny, M. L., & Dean, R. S. (1986). Interspersed post passage ques-
tion and reading comprehension achievement. Journal of Educational
Psychology, 78, 228-229. doi:10.1037/0022-0663.78.3.228
Spooner, A. L. R., Gathercole, S. E., & Baddeley, A. D. (2006). Does
weak reading comprehension reflect an integration deficit? Journal
of Research in Reading, 29, 173-193.
doi:10.1111/j.1467-9817.2006.00284.x
Sundbye, N. (1987). Text explicitness and inferential questioning: Effect
of story understanding and recall. Reading Research Quarterly, 22,
82-98. doi:10.2307/747722
Tal, N. F., Siegel, L. S., & Maraun, M. (1994). Reading comprehension:
The role of question type and reading ability. Reading and Writing, 6,
387-402. doi:10.1007/BF01028850
Van den Broek, P., Rohleder, L., & Narvaez, D. (1994). Cognitive proc-
esses in the comprehension of literary texts. In H. van Oostendorp
and R. Zwaan (Eds.), Naturalistic Text Comprehens ion (pp. 229- 246).
Norwood, New Jersey: Ablex Publishing Corp.

P. KÄNGSEPP 387
Van den Broek, P., Tzeng, Y., Risden, K., & Trabasso, T. (2001). In-
ferential questioning: Effect on comprehension of narrative text as a
function of grade and timing. Journal of Educational Psychology, 9,
521-529. doi:10.1037/0022-0663.93.3.521
Van Dijk, T. A., & Kintsch, W. (1983). Strategies of discourse com-
prehension. New York: Academic Press.