Content uploaded by Anila Fejzo
Author content
All content in this area was uploaded by Anila Fejzo on Jan 21, 2016
Content may be subject to copyright.
Title: The contribution of morphological awareness to the spelling of morphemes and
morphologically complex words in French in students from low socioeconomic
backgrounds
Abstract: The goal of this study was to explore the relationship between morphological
awareness and the spelling of morphemes and morphologically complex words among
third- and fourth-grade Francophone students. To reach this objective, a dictation
including morphologically complex words with prefixes, roots, morphogrammes and
suffixes was given. The targeted items had inconsistent or infrequent spellings, so their
production would require morphological knowledge. The seventy-five participants, from
a low socio-economic background, also completed three tests that measured
morphological awareness. Correlational analyses indicated that the explicit level of
morphological awareness was significantly correlated with the spelling of each type of
morpheme, but regression analyses show that it makes a unique contribution only to the
spelling of morphogrammes (4%), suffixes (9%), and to the production of
morphologically complex words (5%) after grade level, word identification, non-verbal
intelligence and phonological awareness were partialled out. Results corroborate those of
previous studies and support the hypothesis that morphological awareness plays an
important role in the production of morphologically complex written words. Moreover,
these findings contribute to the discussion of the nature of the link between
morphological awareness and word spelling.
Introduction
Writing in a morphophonemic language, like French, requires the use of knowledge
related to morphology, among others. Morphological awareness refers to the abilities and
knowledge that allow for reflection on and manipulation of word structure (Carlisle
(1995). These abilities have been shown to be related to the spelling of words and
morphemes. In order to verify this relationship, many studies have measured the spelling
of morphemes comprised of phonemes which have multiple corresponding graphemes.
Since the choice of the correct grapheme requires morphological knowledge, the level of
1
morphological awareness is correlated with the correct spelling of targeted morphemes
and words. In some studies, morphological awareness contributed to the spelling of
words through morpheme spelling; consequently, the hypothesis of specific relationship
between the two was posited (Nunes, Bryant, & Bindman, 1997; Sénéchal, Basque, &
Leclaire, 2006). However, in other studies, where morphological awareness influenced
the spelling of words even that of monomorphemic ones (Casalis, Deacon, & Pacton,
2011; Deacon, Kirby, & Casselman-Bell, 2009; Sangster & Deacon, 2011), the
hypothesis of a general relationship between the two seems to merit further investigation.
Since these studies targeted one morpheme at a time, it would be interesting to verify the
contribution of morphological awareness to the spelling of different types of morphemes
and the morphologically complex words that they compose. The comparison of their
respective contribution will help to understand the nature of the relationship between
morphological awareness and the spelling of morphemes and morphologically complex
words.
Morpheme encoding in written French
The role of morphological awareness in the spelling of written words comes from the
way in which words are encoded in a given language. In French, like in all alphabetical
languages, there are two levels of encoding the words (Jaffré, 1999). In the first level of
encoding, which concerns the phonological form of a word, a letter or a group of letters,
called a grapheme transcribes each phoneme of the word. An equal number of phonemes
and graphemes in a language would lead to transparent encoding, which is the case for
Albanian where 36 graphemes are used to transcribe 36 phonemes. On the other hand, a
disparity between the number of phonemes and the number of graphemes leads to opaque
encoding, which is the case of French where 130 graphemes are used to transcribe 36
phonemes (Catach, 1978). The lack of biunivocity is at the origin of the inconsistency
between graphemes and phonemes (Ziegler & Goswami, 2005) which happens to be
bidirectional in French. On the one hand, the same grapheme corresponds to different
phonemes (e.g. the grapheme t corresponds to /t/ ou /s/ like in tiers (third) and
démocratie (democracy)). But this kind of inconsistency is low in French, (less than 0.2%
2
of words encountered by readers during the first years of primary school (Peereman,
Lété, & Sprenger-Charolles, 2007). On the other hand, the same phoneme is transcribed
by multiple graphemes (e.g. the same /ã/ is realized by an, am, en, etc. like in
embarrassant (embarrassing)). This inconsistency concerns 82% to 88% of the
phoneme-grapheme mappings in written words encountered by readers from first to third
grade (Peereman et al., 2007). It should be noted that not all of alternatives are equal as
some graphemes are easier to transcribe than others. In fact, Francophone spellers have
less difficulty transcribing contextually consistent graphemes (e.g. gu for /g/ followed by
e like in guerre (war)) than inconsistent ones (e.g. ff for /f/ like in affaiblissement
(weakening)) (Mousty & Alegria, 1999).
If this first level of encoding is characterized by inconsistency, which makes written
French difficult to master (Jaffré & Fayol, 2006), the second level, that of meaning, is
generally consistent. This consistency is reflected in the stable transcription of
meaningful units, words and morphemes. For example, in the sentence Une goutte d’eau
ne goute pas comme une goutte d’huile (A drop of water does not taste like a drop of oil),
goutte is written the same way twice since it refers to the same word, whereas goute (to
taste), while having the same phonological form, is written differently as it refers to a
different meaning. Encoding meaning in French is guided by the morphographic principle
(Jaffré, 1999), which consists of a stable transcription of morphemes, which can be
prefixes, roots and suffixes (e.g. the root lait (milk) is written the same way in words in
its morphological family: lait-ier (milkman), a(l)-lait-er (breastfeed), laitage (dairy),
etc.). The inflexional aspect of this principle allows the consistency of grapheme-
phoneme mappings in final-word position to be increased by 11% (Peereman, Sprenger-
Charolles, & Messaoud-Galusi, 2013). For example, the overall consistency of the /ã/
phoneme is .429. When morphological information is taken into account (e.g. this
phoneme as the ending of an adverb) its consistency increases to 1 (the grapheme is
always –ent) (ibid).
The derivational aspect of this principle governs, among other things, 1) the conservation
of the same graphemes in the roots of words in the same morphological family (e.g. terr-
and not taire (shut) in the family composed from the root terre (land): a(t)terrir (to land),
3
enterer (to bury), déterrer (to dig out), etc.) 2) the conservation of the same graphemes in
prefixes and suffixes (e.g. the prefix en- (in-) or the suffix –ment (-ly) always has the
same orthographic form, never an-, ant or man, mend, mant, etc.), 3) the conservation of
morphogrammes or mutogrammes1, silent letters at the end of a word that are pronounced
in words from the same morphological family (e.g. t in début (beginning) like in débuter
(to begin), debutant (beginner)), 4) the presence of double consonants at the borders of
morphemes (e.g. innombrable (innumerable) between the prefix and the root or
abandonner (to abandon) between the root and the suffix).
These morphological regularities can help the speller to counterbalance the effect of
graphemic inconsistency stemming from the first level of encoding. In fact, a speller who
knows these regularities will choose an infrequent grapheme without difficulty when it is
a part of a morpheme (e.g. em in embellir (to embellish), instead of an which is the
dominant grapheme). If morphological information is expected to increase the graphemic
consistency (Peereman et al., 2013), how can we explain the fact that morphological
encoding is a source of difficulty for spellers (Jaffré et Fayol, 2006)? Is it the fact that
morphological encoding is not completely regular? For example, the prefix a- sometimes
is followed by a double consonant like in atterrir (to land) and sometimes not such as in
alourdir (to weigh down) or, a morphogramme (e.g. t in pivot (pivot)) can be guessed by
using another member of its morphological family (e.g. in pivoter (to pivot)), but the
Francophone speller can be misled and spell clout instead of clou (nail) relying on clouté
(studded). Or are there other factors related to a speller’s morphological information
knowledge that are at the root of these difficulties? To understand the contradiction
between real and expected effects of the second level of transcription on the spelling of
written words, it seems necessary to adopt a psycholinguistic perspective since these
effects are difficult to explain from a linguistic perspective. To this end, models of
orthographic development and empirical data are presented in the next section. This
1 In order to facilitate reading of this paper, morphogrammes are listed among the
targeted morphemes in the rest of the text, even though they are not morphemes.
4
presentation will allow for a better comprehension of how morphological awareness
intervenes in the spelling of morphologically complex words.
Morphological awareness and the spelling of morphologically complex words
In a transparent orthography, the spelling of morphologically complex words can be the
result of a strong ability to segment words phonologically and a good knowledge of
phoneme-grapheme correspondences (e.g. faleminderit (thank you) in Albanian). In an
opaque orthography, these abilities are not sufficient for the spelling of all words, as it is
necessary to retrieve orthographic representations of words encoded in the mental lexicon
(Fayol, 2008). Establishing these representations in the mental lexicon, as described in
models of orthographic development (Ehri, 1999; Frith, 1985; Seymour, 1999), requires
mastering the knowledge of correspondences between phonemes and graphemes as well
as a good level of phonological development. Through multiple encounters with written
words, the speller gradually constructs orthographic representations of longer and longer
sequences of letters. Their storage in the mental lexicon is reinforced and their retrieval is
facilitated by connections with their respective phonological and semantic representations
(Ehri, 2014). Of the sequences of letters that have both a meaning and a phonological
form, morphemes are ideal candidates to be represented in the mental lexicon and to be
retrieved by the speller. Beside their orthographic, phonological and semantic forms,
morphemes are frequentally encountered by readers/writers because they form a part of
many derived or inflected words. In this way, when spellers see the morphologically
complex word encerclement (encirclement), they create a representation in their mental
lexicon or reinforce existing orthographic representations of the morphemes which
compose it, such as the prefix en-, the root –cercle- (circle) and the suffix –ment. Yet the
creation of these representations presupposes abilities to reflect on and to manipulate
morphemes within words, aside from a good level of phonological awareness, the
knowledge of the phoneme-grapheme correspondences and frequent encounters of words
that are constructed with these morphemes. Difficulties in developping these abilities, in
other words a low-level of morphological awareness, could compromise the storage of
orthographic representations of morphemes (Seymour, 1999) and cause errors when
spelling morphologically complex words. For example, a speller with deficient
5
orthographic representations because of a low-level of morphological awareness could
mistakenly write ansercleman (for encerclement) using dominant graphemes.
Empirical data support the hypothesis put forward by developmental models that
morphological awareness contributes to word spelling (Bryant, Nunes, & Bindman, 2000;
Casalis et al., 2011; Kemp, 2006; Nunes, Bryant, & Bindman, 2006; Sangster & Deacon,
2011; Sénéchal, 2000; Sénéchal et al., 2006; Treiman & Cassar, 1997). Rubin (1988) and
Treiman and Cassar (1997) observed that preschool children start developing
morphological knowledge and they use it when they spell. Participants had a tendency to
omit a grapheme (e.g. « n ») in a monomorphemic word (e.g. brand) and to keep it in a
bimorphemic word (e.g. rained).
Similar results were observed in a population of older students when other morphemes
were targeted. Students in second, third and fourth grade found it easier to produce
complex graphemes when they formed part of the root of the word; for example, silent
graphemes (e.g. k in know), multigramme graphemes (e.g. aught in naughty) and unusual
correspondences between phonemes and graphemes (e.g. s in treasure) (Nunes et al.
2006). Morphological awareness was shown to make a unique contribution to the spelling
of words containing complex graphemes. Furthermore, Sangster and Deacon (2011)
asked students between the ages of 6 and 10 to complete bimorphemic (e.g. lucky) and
monomorphemic words (e.g. study) by circling the final grapheme from three options,
one of which was a suffix. From age 9 onwards, participants were better able to write the
final graphemes when they corresponded to suffixes than they were a part of a
monomorphemic word. The production of the correct graphemes was correlated with
morphological sensitivity, which seemed to be substantially put in place by the fourth
year.
The results of studies carried out in French context also corroborate the hypothesis that
morphological awareness is used when spelling words. As early as the third grade, French
spellers found it easier to choose the correct grapheme from the alternatives to transcribe
a phoneme when it was a part of a plurimorphemic word (e.g. laitier) (milkman) than
when it was a part of a monomorphemic word (e.g. falaise) (cliff) (Casalis et al. 2011).
6
Morphological awareness was correlated with their performance in word spelling.
Sénéchal (2000) and Sénéchal et al. (2006) obtained similar results with another type of
problematic letter in French, morphogrammes. The researchers targeted items where the
final silent consonant could be revealed by other words belonging to the same family
(morphological item, e.g. fusil, fusiller) (gun, shoot) or not (lexical item, e.g. foulard)
(scarf). Results of these two studies showed that fourth-grade students performed better
with morphological items than with lexical items. What is more, students who had
developed a good level of morphological awareness had a tendency to develop a
morphological strategy that allowed them to correctly produce morphological items or
those containing morphogrammes. In the same vein, Casalis (2003) found convincing
results pertaining to the contradictory effects of morphological encoding. In this study,
fourth-grade students had to spell monomorphemic and bimorphemic words containing
both consistent graphemes and rare, inconsistent ones. The latter are found in roots (e.g.
cent in centenaire) (centenarian), suffixes (e.g. eau in renardeau) (young fox) or at the
border of morphemes (e.g. c in lionceau) (lion cub). The difference in performace
between advanced spellers and less advanced ones was greater for the bimorphemic
words than for the monomorphemic words. If advanced learners benefited from the
presence of morphemes in bimorphemeic words, less advanced learners were
disadvantaged by the presence of these same morphemes.
Taken together, these studies support the conclusion that morphological knowledge can
be beneficial when spelling morphologically complex words with inconsistent
graphemes. Furthermore, the developmental level of morphological awareness appears to
be linked to the choice of these graphemes; a high level of morphological awareness
seems to facilitate it and a low level of morphological awareness seems to disadvantage
learners in the spelling of morphologically complex words. If this link explains the
contradictory impact of the second level ecoding on word spelling, we need to know how
this link is established among Francophone spellers.
7
The present study
Our primary objective is to examine how morphological awareness is related to the
spelling of each type of morpheme. To reach this objective we chose a population of
third- and fourth-grade students, who, according to the previously described
developmental models, are in the phase where they establish representations of
morphemes. In this same vein, we targeted morphologically complex words which have
been shown to be difficult for the population of our study (according to the EOLE, third-
grade students spell the targeted items accurately less than 80% of the time (Pothier &
Pothier, 2004). We assume that this difficulty comes from inconsistent or rare graphemes
which are found in a) prefixes, b) bases, c) morphogrammes and d) suffixes. Finally, to
better isolate the contribution of morphological awareness in the spelling of morphemes
and morphologically complex words, we measured predictor variables for word spelling
in light of the described developmental models and other studies similar to ours (Deacon
et al. 2009; Nunes et al. 1997; Casalis, 2003). Therefore, we measured phonological
awareness, which is a prerequisite for establishing orthographic representations. Indeed,
Sprenger-Charolles, Siegel, Béchennec, and Serniclaes (2003) found that phonological
processing contributes to orthographic representations among French fourth graders. In
measuring word identification, we controlled for knowledge of grapheme-phoneme
correspondences, and written word exposure, two essential conditions for the storage of
orthographic representations. Finally, it was necessary to measure cognitive deductive
capacities necessary for manipulating sequences of letters that form morphemes and for
discovering spelling regularities that influence word production. We therefore
administered the Raven Matrices, a non-verbal intelligence test. We expect that
morphological awareness will contribute to morpheme spelling, even after controlling for
these three variables.
Our second objective was to verify the contribution of morphological awareness to the
spelling of morphologically complex words and to understand the relationship between
the two. Does morphological awareness make a unique contribution to the spelling of
morphologically complex words after controlling for morpheme spelling? If, after this
factor is partialled out, morphological awareness retains its contribution to the spelling of
8
morphologically complex words, this would mean that their relationship is general.
However, if this is not the case, the absence of such a contribution would suggest a
specific relationship between morphological awareness and word spelling.
Method
Population
Forty-three third-grade students (mean age 8.7 years, SD= 0,43) and 32 fourth-grade
students (mean age 9.8 years, SD=0,56) from a school in a low SES neighbourhood in the
Montreal region participated in the study (index of SES of school being 9 where 10 is
attributed to the most disadvantaged schools (MELS, 2014)). According to the
questionnaire given to parents, 72% of the students speak French at home. The other
students (speaking English (7), Espagnol (4), Roumanian (2), Chinese (2), Arabic (1),
Bengali (1), Bosnian (1), Creole (1), Punjabi (1) or Vietnamese (1)), even though the
French was not the dominant language at home, were born in Canada. All of their
schooling has been in French and they are fluent in this language. In order to verify if this
linguistic variable has an impact on the participant performances, a test t for two
independent groups was applied to all the measures. As no significant difference was
observed between the two linguistic groups, the linguistic variable will not be taken into
account in the results or the discussion.
Materials
Morphological awareness
An experimental test that included three tasks was used to measure morphological
awareness. The first two required participants to complete a sentence with a derived word
(e.g. Celui qui dit des choses vraies dit la … (vérité)) (The one that says true things says
the … (truth)) or with the root of a derived word (e.g. Dans ce conte féerique, l’un des
personnages est la … (fée) du boisé) (In this fairytale, one of the characters is the …
(fairy) of the woods) which was presented in bold font in the first part of the sentence.
The items in the two tests were divided into two phonological conditions: neutral (e.g.
9
brulé- brulure) (burned-burn) and non-neutral (e.g. vraie-vérité) (true-truth) following
Fowler and Liberman (1995).
In the third test, inspired by that of Casalis and Louis-Alexandre (2000), participants
were required to derive non-words (e.g. Quand on toulle, on fait un… (toullage ou
toullement)) (When one toulles, he is doing a … toullement). This test targeted the
derivational aspect of morphological awareness because of its strong correlation with
word spelling (Fowler & Liberman, 1995). Each correct answer was worth one point.
Inter-item reliability of the test was a= .894 for the total number of items.
Morpheme and word spelling
For this test, 31 morphologically complex words were dictated in a gap-fill exercise (see
Appendix A). The words were chosen based on their morphological complexity. They
contained 1) prefixes (e.g. irréparable) (irreparable), 2) roots (e.g. encercler) (encircle),
3) morphogrammes (e.g. début) (beginning), 4) and suffixes (e.g. gagnant) (winner)
composed of graphemes whose spelling should be facilitated by the contribution of
morphological knowledge. Data were coded on two different levels. Firstly, one point
was given for each word which was spelled correctly (e.g. immobile) (immobile), for a
total of 31 points. This score figures as word spelling in our tables. The second level of
coding targeted individual morphemes within a word (e.g. the spelling of the prefix im in
immobile). A point was given for each morpheme which was spelled correctly, even if the
word was not spelled correctly (e.g. immobil). This allows for two or three morphemes to
be coded within the same word (e.g. 3 morphemes in affaiblissement) (weakening).
Consequently, responses were scored out of a total of ten points for each experimental
condition (prefixes, roots, morphogrammes and suffixes) for a total of 40 points
(morpheme spelling score). Inter-item reliability was a=.929 for the total number of
items.
Word identification
The capacity to identify words was measured through the word recognition measure
(MIM) from the BELEC battery (Mousty, Leybaert, Alegria, Content, & Morais, 1994).
10
Test A was administered. This test is comprised of an equal number of short and long
words for a total of 72 words. Long items are plurimorphemic words. The score
pertaining to participant accuracy in word identification was included in our analyses.
Phonological awareness
The phonological awareness test from the BELEC (Mousty et al, 1994) chosen for our
study requires participants to carry out two successive phonological manipulations: first,
isolating the first phoneme of two words and, second, joining these phonemes to create a
syllable. The test is composed of sixteen pairs of words in which there is a mismatch
between the initial phoneme of the word and the phoneme which corresponds to the first
letter of the word (e.g. in the word autobus, (bus) the phoneme /o/, which is heard, is
different from the phoneme /a/, the first letter of the word). Each response which
corresponds to the acronym formed by the initial oral phonemes is worth one point. An
answer which takes into account the spelling of the words is not considered a correct
answer (E.g. /ka/ for Cher Auguste is considered incorrect).
Non-verbal intelligence
This variable was measured using Raven’s Matrices (1983). Participants had to choose
the image which best completes the pattern from a group of six images. Since this test
measures participants’ ability to find patterns, it seemed fitting to use this measure in this
study which is targeting participants’ ability to find morphological patterns, notably in the
spelling of morphologically complex words. Controlling this variable allowed us to better
isolate the contribution of morphological awareness in morpheme and word spelling.
Because of the participants’ age, the first three matrices were given following previous
research (Kirby et al., 2012). Each matrix has 12 exercises which are presented through
images. There is only one correct answer which is worth one point.
Procedure
11
Tests were administered in September. The word and morpheme spelling tests as well as
the morphological awareness test were administered to the class as a group (45 minutes).
At the beginning, the researcher read the entire dictation, and then the text was read
again, pausing at the targeted words. The morphological awareness test was also read by
the researcher in order to counter the potential bias of the participants’ reading skills. One
week later, the phonological awareness and the word identification tests were
administered individually in a quiet room in the school. The non-verbal intelligence test
was administered in groups of 7 to 9 children, as suggested by the experimental protocol
for this age group.
Results
In order to reach our objectives of examining the relationship between morphological
awareness and the spelling of morphemes and morphologically complex words (for sake
of space, this variable is entered as word spelling in all tables), we carried out various
quantitative analyses: descriptive statistics (Table 1), correlational (Table 2 and Table 3)
and regression analyses (Table 4 and Table 5).
In Table 1, means, standard deviations and 95% confidence intervals are presented for all
variables of the study. The results of the spelling of each of the morphemes showed that
our participants had difficulty with the spelling of morphemes, specifically with prefixes
and roots. Morphogrammes were better spelled by third graders and suffixes were the
best morpheme spelled by fourth graders.
Tabel Means, Standard Deviation and 95% Confidence Intervals for Morphological Awareness and
Spelling tasks
Third grade
(n=43)
Fourth grade
(n=32)
Measures
M (SD) 95% CI M (SD) 95% CI
Prefix spelling 16.28 (18.13) [11.16, 22.09] 36.88 (19.58) [30.63, 43.43]
Base spelling 18.60 (16.70) [13.95, 23.72] 40.63 (18.65) [34.38, 47.19]
Morphogramme spelling 35.58 (29.05) [27.44, 45.11] 63.13 (24.16) [55.00, 70.94]
Suffix spelling 26.98 (21.33) [21.16, 33.25] 63.44 (23.63) [55.31, 70.94]
Word spelling 17.90 (16.91) [13.26, 23.18] 39.37 (16.53) [34.06, 44.68]
Word derivation 41.86 (27.54) [33.72, 50.47] 58.13 (21.77) [50.63, 65.63]
12
Base extraction 36.05 (24.60) [28.61, 43.72] 58.13 (21.62) [50.63, 65.93]
Non-word derivation 10.00 (11.34) [6.75, 13.72] 24.06 (19.32) [17.81, 30.94]
Short word identification 85.34 (11.74) [81.78, 88.76] 89.83 (6.51) [87.48, 91.98]
Long word identification 67.44 (18.12) [61.69, 72.74] 76.13 (12.38) [71.53, 80.21]
Non-verbal intelligence 23.88 (4.14) [22.60, 25.16] 27.68 (4.98) [25.84, 29.25]
Phonological awareness 71.22 (22.34) [63.95, 77.76] 73.83 (13.88) [68.95, 78,52]
With a view to better understanding our results, we verified the effect of frequency and
word length, factors known to be significant predictors in word spelling (Mousty &
Alegria, 1999). These factors were not controlled when items were selected as other
criteria were taken into consideration (see Method; Morpheme and word spelling). Our
analyses revealed that neither frequency nor word length, measured by the Lexique
database (New & Pallier, 2001) influenced morpheme spelling. However, when the data
were divided according to the type of morpheme, we observed that frequency was
significantly correlated (r=.66, p=.03) with the spelling of suffixes in grade 4.
Furthermore, when results on spelling of morphologically complex words were analysed,
they indicated that participants in both grades performed significantly poorly compared to
standardized scores (Pothier & Pothier, 2004). In fact, the one-sample t test showed that
the difference was significant for both grades (t=9.95, p= .000 for third graders and
t=6.625, p= .000 for fourth graders) the standardised performance on these words being
43.58% in grade 3 and 58.74% in grade 4.
Despite the low performance, findings showed that fourth graders significantly
outperformed third graders (t=5.49, p =.000) on word spelling involving inconsistent
graphemes choice converging with the findings of a longitudinal study among French
spellers (Alegria & Mousty, 1996). In this study, a significant improvement was observed
at the end of the third year when compared with scores pertaining to the spelling of
inconsistent graphemes at the end of second year.
In addition, students at both levels had weak results on the morphological tests, notably in
the test targeting non-word derivation. This is not surprising insofar as this task requires a
very high level of morphological awareness (Colé, Royer, Leuwers, & Casalis, 2004).
Thus, this population seems also to have a low level of morphological awareness too.
However, fourth graders performed significantly better than third graders on all measures
of morphological awareness (t=4.05, p=.000).
13
Results on predictor variables show that our participants perform below standardised
scores. At the end of 2nd grade, according to standardised scores (Mousty & Laybaert,
1999) French readers should perform at 86,84% on short words identification and 71,06%
on long word identification. The third graders in the present study scored respectively
85.34% and 67.44% on these measures. Our participants in grade 4 also had lower results
than the standardised scores for third graders (89.93 versus 94.73 for short words and
76.13 versus 86.64 for long words). The difference of the performance in this grade with
standardised scores was significant for both measures (respectively t= 4.262, p= .000 and
t= 4.803, p= .000). It is noteworthy that the difference with standardised scores is bigger
for long word identification for all participants.
A deeper gap is observed for phonological awareness; third graders as well as fourth
graders scored lower than standardised scores (respectively 71.22% versus 79.9% and
73.83% versus 85.8%)(Mousty & Laybaert, 1999). The difference with standardised
scores was significant for both grades (respectively t=2.54, p=.015 and t=4.879, p=.000).
Likewise, the score obtained in non-verbal intelligence (23.88 in grade 3) place third
graders below the 50th percentile but above the 25th percentile for their age (the
standardised scores being respectively 31 and 22 (Raven, 2000)). Fourth graders are
situated in the 25th percentile with their score of 27.68 (28 being the standardised score
for this percentile).
To summarize the descriptive data, participants have consistently low scores on all
measures. Their scores are significantly lower (except for third graders on word
identification) than standardised scores. These results add to existing data supporting the
hypothesis of a relationship between students’ academic achievement and their
socioeconomic background (Sirin, 2005). Consequently, it would be more prudent to
place the findings of this study in the context of an academically disadvantaged
population.
On the other hand, the results of the correlational analysis (Table 2) yield two findings:
first, correlations between morphological tasks and morpheme spelling are a function of
grade level. In grade 3, morpheme spelling for all types of morphemes were correlated
with all morphological tasks, except for prefix spelling which is correlated only with the
non-word derivation task. This task is the only morphological task to be correlated with
14
all types of morpheme spelling in grade 4. Second, when linked with the results presented
in Table 1, these correlations reveal that the weaker the results are on the morphological
tasks, the more they are correlated with the spelling measures. In contrast, the weaker the
results are in morpheme spelling, the less they are correlated with the morphological
tasks.
Tabel Correlations between Measures on Morphological Awareness and Morphem and Word Spelling
1 2 3 4 5 6 7
1. Word derivation _ .72** .59** .18 .29 .05 .25
2. Base extraction .73** _ .55** .12 .29 .07 .30
3. Non-word deriv. .63** .64** _ .42* .35* .37* .48**
4. Prefix spelling .19 .25 .45** _ .37* .60** .37*
5. Base spelling .39** .47** .48** .61** _ .62** .55**
6. Morphogram sp. .33** .54** .50** .74** .77** _ .66**
7. Suffixe spelling .35** .47** .61** .66** .75** .82** _
** p < .01. * p < .05. Correlations for third graders are presented below the diagonal. Correlations for
fourth graders are presented above the diagonal.
To continue our analysis, we verified the correlations between predictor variables and the
spelling of each type of morpheme and the spelling of morphologically complex words.
We decided to calculate them by entering the entire sample because of the small number
of participants. As there is a significant difference between third and fourth graders, grade
was entered as a predictor in the further analyses. For morphological awareness we
entered the score of non-word derivation task, as this was the only task to be correlated
with the spelling of each type of morpheme.
Table Correlations between Predictor Variables and Morpheme and Word Spelling
15
1 2 3 4 5 6 7 8 9 10
1. Grade -
2. Short word identification .22 -
3. Long word identification .26* .65** -
4. Non-verbal intelligence .38** .28* .39** -
5. Phonological awareness .07 .36** .54** .48** -
6. Morphological awareness .42** .28* .40** .31** .27* -
7. Prefix spelling .48** .34** .51** .15 .19 .45**
8. Base spelling .53** .45** .50** .34** .25* .48** .61** -
9. Morphogramme spelling .45** .53** .55** .37** .29* .50** .74** .77** -
10. Suffix spelling .63** .54** .52** .39** .26* .61** .66** .75** .82** -
11. Word spelling .54** .53** .60** .40** .28* .57** .79** .85** .95** .91**
** p < .01. * p < .05.
The results of the correlational analysis presented in Table 3 show that all predictor
variables are significantly correlated, except for those between prefix spelling and non-
verbal intelligence and phonological awareness. Furthermore, morphological awareness
was moderately correlated with each of the targeted morpheme spelling and word
spelling. The strong correlation between word spelling and morpheme spelling is
expected as difficulty spelling morphemes leads directly to difficulty in spelling
morphologically complex words composed of these morphemes.
Furthermore, if, at first glance, the weak correlations between phonological awareness
and word and morpheme spellings are surprising, they were foreseeable given the items
to be spelled. Indeed, we chose items that when spelled, using phonological awareness
and grapheme-phoneme correspondences was not sufficient. The spelling of these items
requires the choice of a rare grapheme among multiple alternatives and consequently, this
diminishes the contribution of phonological awareness.
The next step in the regression analyses was to enter predictor variables into the
hierarchical regression model in descending order in terms of their correlation with
morpheme spelling (Howell, 2008): grade, long word identification, short word
identification, non-verbal intelligence and phonological awareness (see Tables 4 and 5).
We calculated the predictive value of all these variables in step 1. When morphological
awareness was entered into the equation in step 2, the predictive value of the model
increased for each morpheme spelling and we present the R2 change for each morpheme.
16
This change represents the unique contribution of morphological awareness to the
morpheme spelling after having controlled for the variables in the first model.
Table Hierarchical Multiple Regression Analyses Predicting each Morpheme Spelling
Morpheme spelling
Prefix spelling Base spelling Morphogramme
spelling
Suffix spelling
Predictor ∆R2ß ∆R2ß ∆R2ß ∆R2ß
Step 1 .39** .44** .45** .57**
Grade .32** .41** .32** .54**
Long word identification .36** .25 .29* .38**
Short word identification .02* .19 .27* .42**
Non-verbal intelligence .03 .07 .04
Phonological awareness .00 .02 .02
Step 2 .02 .03 .04* .09**
Morphological awareness .17 .20†.23* 42**
Total R2.41** .47* .49** .66**
n75 75 75 75
† p=.054. *p < .05. ** p < .01.
Our findings show that when morphological awareness was entered into this conservative
regression model (Howell, 2008), the predictive value increased for each targeted
morpheme, but the R2 change was significant only for the suffixes and morphogrammes.
According to these data, morphological awareness has a significant unique contribution
for suffix (9%) and morphogramme (4%) spelling, but its unique contribution to prefix
(2%) and base (3%) spelling is not at a significant level.
To reach our second objective, we carried out the same regression analysis to verify the
contribution of morphological awareness to word spelling and we found that the factors,
taken together, predicted 55% of the variation in word spelling. After adding
morphological awareness into the regression model, the predictive value of all variables
increased significantly to 60%.
Table Hierarchical Multiple Regression Analyses Predicting Morphologically Complex Word Spelling
Morphologicaly complex word spelling
Without morpheme spelling With morpheme spelling
Predictor ∆R2ß ∆R2ß
Step 1 .55** .98**
Grade .40** -.08**
Long word identification .35** .04
Short word identification .21* -.01
17
Non-verbal intelligence .06 .08**
Phonological awareness -.02 -.05
Morpheme spelling 1.00**
Step 2 .05** .00
Morphological awareness .27** .00
Total R2.60** .98**
n75 75
* p < .05. ** p < .01.
The unique contribution of morphological awareness on morphologically word spelling is
5% at a significant level (p=.003) after controlling for grade, long and short word
identification, non-verbal intelligence and phonological awareness. With regard to the
unique contribution of morphological awareness to word spelling, a morpheme spelling
score was calculated by summing the scores of each morpheme. The unique contribution
of morphological awareness to morpheme spelling was found to be significant (p=.003)
and equalled that to word spelling, that is 5%. This outcome suggested that
morphological awareness may no longer contribute to word spelling if morpheme
spelling is partialled out. Indeed, our regression analysis indicated that when morpheme
spelling is entered into the regression model in step 1, morphological awareness no
longer has a predictive value for word spelling (see Table 5).
Discussion
Our first objective was to examine the relationship between morphological awareness and
morpheme spelling in third and fourth grade students. In order to meet this objective, we
targeted four types of morpheme: prefixes, bases, morphogrammes and suffixes, which
form part of morphologically complex words. The results of the descriptive analyses
suggest that our participants encountered great difficulty in spelling the targeted
morphemes. The scores on spelling prefixes and bases were lower than those on the
spelling of morphogrammes and suffixes. Furthermore, the results of the morphological
tests indicated that the level of development of their morphological awareness was weak.
The correlational analysis showed that each morphological awareness task and the
spelling of each type of morpheme spelling were significantly correlated in third grade,
except for prefix spelling. In grade 4, however, the non-word derivational task was the
only morphological task that was significantly correlated with the spelling of each type of
18
morpheme. Regression analyses indicated that the predictive value of variables for the
spelling of each type of morpheme increased when morphological awareness was entered
into equation in step 2. But this increase was significant only for the spelling of suffixes
and morphogrammes. More specifically, morphological awareness made a unique
contribution of 9% to suffix spelling and one of 4% to morphogramme spelling after
controlling for grade level, word identification, phonological awareness and non-verbal
intelligence. These findings add to results of other studies which also observed a unique
contribution of morphological awareness to morphogramme spelling (Sénéchal et al.
2000; 2006) and extend those of Sangster and Deacon’s study (2011) suggesting that by
the age of 9 students are sensitive to the relationship between suffix spelling and
morphological information.
A detailed analysis of the results obtained in the present study can help to refine our
understanding of this contribution. First of all, putting together the descriptive statistics
(Table 1) and the correlations (Table 2) we found that the weaker the results on the
morphological tests, the more they are correlated with morpheme spelling on the items
with the highest number of correct scores (e.g. correlations between results for suffix
spelling and the non-word derivation task). This leads us to presume, on the one hand,
that the more a morphological task requires an explicit level of morphological awareness,
the stronger its relationship will be with morpheme spelling. Indeed, the non-word
derivation task, which measures the most explicit level of morphological awareness (Colé
et al., 2004) was significantly correlated with all types of morpheme spelling for both
grade levels which was not the case for the word derivation and base extraction tasks. It
seems as though students who have developed an explicit level of morphological
awareness are more likely to improve their representation of morphemes, and those who
did not develop their morphological skills experience difficulty in correctly spelling
morphemes. In the same vein, Rubin (1988) concluded that it is mainly difficulties at the
explicit level of morphological awareness that can explain difficulties in spelling
inflected and derived words. However, this interpretation should be treated with caution
given the effect of word frequency on the spelling of suffixes for fourth grade students.
On the other hand, it seems that the lack of unique contribution of morphological
awareness to prefix and base spelling is due to the low scores of our participants on these
19
items. As our participants came from a low socio-economic background and performed
poorly in all tasks, it is then relevant to suppose that the contribution of morphological
awareness would be larger in less disadvantaged populations. Further research is needed
to elucidate this hypothesis.
Moreover, the significant unique contribution of morphological awareness on suffix and
morphogramme spelling is particularly eloquent when statistical data about phoneme-
grapheme mappings (Peereman et al., 2007) are taken into account. Phoneme-grapheme
mappings in final position to which French spellers are exposed during the first years of
school, are the most inconsistent (55% versus 25% in middle position and 9% in initial
position). This inconsistency, due to the derivational characteristics of French
orthography (ibid) incites the speller to pay attention to final phoneme-grapheme
mappings. When these inconsistent mappings have morphological value, which is the
case of morphogrammes and suffixes, they are easier to store in memory, providing that
the speller has discovered the morphological links. Frequent encounters and an explicit
level of morphological awareness are therefore necessary to correctly spell suffixes and
morphogrammes. In fact, the fact that suffix spelling was correlated with their frequency
in grade 4 goes in this direction. In addition, because they are less exposed to
inconsistent phoneme-grapheme mappings in initial (9%) and middle position 25%
(Peereman et al. 2013), French spellers pay less attention to these inconsistent mappings,
and consequently, they do not discover their morphological value. They thus have
difficulty spelling them. According to that speculation, morphological awareness and
spelling of morpheme have a circular causality relationship (which is synonym term of
bidirectional causality relationship (Nunes et al. 2006)). It supports the hypothesis
according to which increasing experience with written language serves as a catalyst for
the development of metalinguistic capacities (Gombert, 1990), notably morphological
ones (Chomsky, 1970). Our explanation is consistent with orthographic representation
construction as it is postulated by orthographic developmental models (Ehri, 2005;
Seymour, 1997). However, studies that investigate the spelling of inconsistent graphemes
pertaining or not to morphemes, after their frequency and position in word are controlled,
are necessary to confirm this hypothesis.
20
Next, taking into account the fact that neither frequency nor word length influenced
morpheme spelling, except for suffixes in fourth grade, it is pertinent to consider that the
morphological abilities and knowledge necessary to correctly spell targeted morphemes
are not the same. In fact, the results obtained with the morphogrammes that have the
highest success rate in third grade also suggest that their spelling depends on the type of
morphological knowledge required. To produce this type of grapheme correctly, spellers
need to understand that the word containing the morphogramme belongs to the same
morphological family as the other words to which he needs to refer in order to choose the
final silent consonant (e.g. d in tard, tardif, retarder, etc.). In fact, this knowledge that
Tyler and Nagy (1987) call relational knowledge, is expected to be developed prior to
other morphological knowledge (syntactic and distributive) and to be well established at
the age of our participants (Tyler and Nagy, 1987). What is more, the fact that the third
graders’ capacity to spell bases was correlated with their performance on the base
extraction task clearly illustrates the relationship between morphological awareness and
morpheme spelling. This correlation suggests that in order to be able to correctly spell the
base of a word, students must be able to find the base in the word. If we take into account
our participants’ weak scores on both tasks, it seems obvious that difficulties extracting
the base from the word may have impeded them from constructing representations of
these types of morphemes. In other words, if participants could not find the correct
grapheme when spelling the base, it could be due to the fact that they are not aware that
this part of the word is the same as in other words that they already know. This
interpretation is even more plausible considering the fact that participants were not able
to benefit from seeing the written form of the six targeted roots presented in the gap-
filling text that they completed during the spelling test (see Appendix A).
Furthermore, the spelling of some of the targeted morphemes required an advanced level
of morphological abilities and knowledge. For example, to correctly spell ir- in
irréparable, the speller not only has to understand that this inconsistent grapheme is the
prefix in, but also has to notice that in certain cases the phonological and orthographic
form of this prefix changes as a function of the initial consonant of the root. In addition,
in order to spell atterrir, students have to discover and apply the regularity of the junction
of the prefix a- above and beyond their morphological analysis skills. Whereas for the
21
spelling of suffixes, except for abandonner which implies a regularity of suffix junction,
the knowledge that a frequently encountered sequence of letters has a meaning and a
phonological form is sufficient to store in the mental lexicon.
In sum, morphological awareness is correlated with the spelling of the four types of
morphemes, but it had a unique contribution only to the spelling of suffixes and
morphogrammes. If statistical analysis allowed for the verification of the unique
contribution of morphological awareness to morpheme spelling, the detailed observations
allowed for a better understanding of their relationship. Our interpretation that a higher
level of morphological awareness is more highly correlated with morpheme spelling and
that difficulty in developing this awareness hinders the establishment of an orthographic
representation of morphemes follows what is proposed in the orthographic developmental
model proposed by Seymour (1997). This model posits that the construction of
orthographic representations of morphemes depends on the level to which morphological
analysis is developed. Moreover, our findings suggest that the variable contribution of
morphological awareness to the different types of morpheme spelling can be attributed to
the type of morphological knowledge implicated and to distributional characteristics of
phoneme-grapheme mappings. Future research is necessary in order to clarify if these
variables explain the relationship between morphological awareness and spelling of
morphemes, when frequency and word length are controlled for. Furthermore, given that
contribution of morphological awareness is significant to the spelling of morphemes that
had better overall success rates and that the population of our study had consistently
lower scores on all measures, especially on the spelling of prefixes and bases, it is
important to verify if a significant contribution can be observed to the spelling of these
morphemes in a population scoring at grade level or above on standardized measures.
In addition, the detailed analysis begs the question of whether measures of morphological
awareness and morpheme spelling are not measures of different levels of morphological
awareness. Indeed, Nunes et al. (2006) report that morpheme spelling “is best described
as an abstract form of morphological awareness” (p.13). It seems nearly impossible to
separate orthographic knowledge from morphological awareness in morpheme spelling.
At the very least, neither current definitions of morphological awareness (Carlisle, 1995;
Kuo & Anderson, 2006), nor models of metalinguistic development (Gombert, 1990;
22
Karmiloff-Smith, 1992) allow for this distinction. This seems to be a promising area to
explore.
Our second objective was to verify the relationship between morphological awareness
and the spelling of morphologically complex words as well as to understand the nature of
this relationship. In order to meet this objective, regression analyses revealed that
morphological awareness had a unique contribution of 5% to the spelling of
morphologically complex words. Since the predictive value of morphological awareness
on word spelling equalled 0 when morpheme spelling was entered into the equation, our
results suggest that morphological awareness contributes to the spelling of
morphologically complex words through morpheme spelling. Our results converge with
that of other studies (Nunes et al. 1997; Nunes et al. 2006; Sénéchal et al. 2006) and
support the hypothesis that there is a specific relationship between morphological
awareness and the spelling of morphologically complex words. At the same time, these
results support orthographic developmental models which posit that representations of
plurimorphemic words are gradually established and that these rely on morpheme
representation (Ehri, 1999; Seymour, 1999).
Conclusion
Our study examined the relationship between morphological awareness and morpheme
spelling as well as the spelling of morphologically complex words among third and
fourth graders in a low socio-economic neighbourhood. It also sought to understand the
nature of this relationship. Our results suggest that morphological awareness is
significantly correlated with the targeted morpheme spelling, but contributes only to
morphogramme and suffix spelling and through this, to the spelling of morphologically
complex words. These results support the hypothesis that there is a specific relationship
between morphological awareness and the spelling of morphologically complex words.
Furthermore, our results demonstrate that third- and fourth-grade students of low SES
have difficulties spelling morphemes and consequently, they have difficulty spelling
morphologically complex words. Five percent of these difficulties can be explained by
participants’ morphological awareness.
23
Nonetheless, certain limitations of our study restrict the extent to which our results can be
generalized. First of all, if frequency and word length of items in the spelling test had
been controlled, we would have been able to better understand the variable performance
in the spelling of different types of morphemes by taking into account the type of
morphological knowledge implicated in their spelling. Second, the way morpheme
spelling and the spelling of morphologically complex words were coded may have
influenced our results, that is to say that it may have masked the contribution of
morphological awareness to word spelling when morpheme spelling was partialled out.
Third, the lack of correlation between two morphological tasks (word derivation and base
extraction) and morpheme spelling in grade 4 begs the question if other than
morphological information (e.g. semantic information) than morphological one was
sufficient to perform these tasks. Thus, it would be cautious to control for this aspect in
developing experimental material in future research.
In spite of this, our results provide precious information to educators about the difficulties
that Francophone third- and fourth-grade students have with regard to morphological
awareness and the spelling of morphemes and morphologically complex words. This
study highlights the contribution of morphological awareness to the spelling of
morphemes and the words that they compose and thus it shows school teachers the
importance of doing activities which foster the development of morphological awareness
in order to better aid students in their spelling development.
References
Alegria, J., & Mousty, P. (1996). The development of spelling procedures in French-
speaking normal and reading disabled children: effects of frequency and
lexicality. Journal of Experimental child psychology, 63(2), 312-338.
Bryant, P., Nunes, T., & Bindman, M. (2000). The relations between children's linguistic
awareness and spelling: the case of apostrophe. Reading and Writing : An
Interdisciplinary Journal, 12(3-4), 253-276.
Carlisle, J. (1995). Morphological awareness and early reading achievement. In L. B.
Feldman (Ed.), Morphological aspects of language processing (pp. 189-209).
Hillsdale: Erlbaum.
24
Casalis, S. (2003). Le codage de l'information morphologique dans l'écriture de mots
chez les apprentis scripteurs. [The morphological information encoding in word
spelling amongst beginning spellers.] Le Langage et l'Homme, 38(2), 95-110.
Casalis, S., Deacon, H., & Pacton, S. (2011). How Specific Is the Connection between
Morphological Awareness and Spelling? A Study of French Children. Applied
Psycholinguistics, 32(3), 499-511.
Casalis, S., & Louis-Alexandre, M.-F. (2000). Morphological Analysis, Phonological
Analysis and Learning To Read French: A Longitudinal Study. Reading and
Writing: An Interdisciplinary Journal, 12(3-4), 303-335.
Catach, N. (1978). L’orthographe. Que sais-je? Paris: Presses universitaires de France.
Chomsky, C. (1970). Reading, Wrting and Phonology. Havard, Educational Review,
40(2), 287-309.
Colé, P., Royer, C., Leuwers, C., & Casalis, S. (2004). Les connaissances
morphologiques dérivationnelles et l'apprentissage de la lecture chez l'apprenti-
lecteur français du CP au CE2. [Derivational knowledge and reading learing in
French students from first to third grade.] Année psychologique, 104, 701-750.
Deacon, S., Kirby, J., & Casselman-Bell, M. (2009). How Robust Is the Contribution of
Morphological Awareness to General Spelling Outcomes? Reading Psychology,
30(4), 301-318.
Ehri, L. (1999). Phases of development in learning to read words. In J. Oakhill & R.
Beard (Eds.), Reading development and teaching of reading : A psychological
persepctive (pp. 79-108). Oxford: Blackwell publishers.
Ehri, L. (2014). Orthographic Mapping in the Acquisition of Sight Word Reading,
Spelling Memory, and Vocabulary Learning. Scientific Studies of Reading, 18(1),
5-21.
Fayol, M. (2008). Apprendre à orthographier la morphologie. [Learning to spell
morphology.] In M. Fayol & J.-P. Jaffré (Eds.), Orthographier [Spelling] (pp.
197-210). Paris: Presses Universitaires de France.
Fowler, C., & Liberman, I. (1995). The role of phonology and orthography in
morphological awareness. In L. Feldman (Ed.), Morphological aspects of
language processing (pp. 157-188). Hillsdale: Erlbaum.
Frith, U. (1985). Beneath the surface of developmental dyslexia. In K. E. Patterson, J. C.
Marshall, & M. Coltheart (Eds.), Surface Dyslexia : Neuropsychological and
Cognitive Studies of Phonological Reading (pp. 301-330). Hillsdale, New Jersey:
Lawrence Erlbaum Associates Publishers.
Gombert, J.-É. (1990). Le développement métalinguistique. [ The metalinguistic
development.] Paris.
Howell, D. (Ed.). (2008). Méthodes statistiques en sciences humaines. [Statistical
Methods for Psychology.] Bruxelles: De boeck.
Jaffré, J.-P. (1999). Orthographe et principes d’écriture en français. [] Rééducation
orthophonique, 200(décembre), 25-34.
Jaffré, J.-P., & Fayol, M. (2006). Orthography and literacy in french. In R. M. Joshi & P.
G. Aaron (Eds.), Handbook of orthography and literacy (pp. 81-104). Mahwah,
New Jersey, London: Lawrence Erlbaum Associates.
Karmiloff-Smith, A. (1992). Beyond Modularity, A Developmental Perspective on
Cognitive Science. Cambridge, Massachusetts, London: The MIT Press.
25
Kemp, N. (2006). Children's spelling of base, inflected and derived words: Links with
morphological awareness. Reading and Writing, 19, 737-765.
Kuo, L.-J., & Anderson, R. (2006). Morphological awareness and learning to read: a
cross-language perspective. Educational Psychologist, 41(3), 161-180.
MELS. (2014). Indices de défavorisation par école 2013-2014. [Socioeconomic index for
schools 2013-2014.]Retrieved january 19 2015 from Ministère de l'éducation du
Loisir et du Sport
Mousty, P., & Alegria, J. (1999). L'acquisition de l'orthographe: Données comparatives
entre enfants normo-lecteurs et dyslexiques. [The acquisition of spelling: data
comparing normal readers and dyslexic children] Revue Française de
Psychologie, 126, 7-21.
Mousty, P., Leybaert, J., Alegria, J., Content, A., & Morais, J. (1994). BELEC : Batterie
d’évaluation du langage écrit et de ses troubles. [Assessment Battery of written
language and its disorders] In J. Grégoire & B. Pierart (Eds.), Evaluer les troubles
de la lecture. Les nouveaux modèles théoriques et leurs implications
diagnostiques [Assessing reading disorders. The new theoretical models and their
diagnostic implications] (pp. 127-145). Bruxelles: De Boeck.
New, B., & Pallier, C. (2001). Lexique 3. [Lexicon]
http://www.lexique.org/docLexique.php
Nunes, T., Bryant, P., & Bindman, M. (1997). Morphological spelling strategies:
developmental stages and processes. Developmental Psychology, 33(4), 637-649.
Nunes, T., Bryant, P., & Bindman, M. (2006). The effects of learning to spell on
children's awareness of morphology. Reading and Writing, 19, 767-787.
Peereman, R., Lété, B., & Sprenger-Charolles, L. (2007). Manulex-Infra: Distributional
characteristics of grapheme-phoneme mappings, and infralexical and lexical units
in child-directed written material Behavior Research Methods, 2007(39), 3.
Peereman, R., Sprenger-Charolles, L., & Messaoud-Galusi. (2013). The contribution of
morphology to the consistency of spelling-to-sound relations: A quantitative
analysis based on French elementary school readers. Année psychologique,
113(1), 3-33.
Pothier, B., & Pothier, P. (Eds.). (2004). Échelle d'acquisition en orthographe lexicale:
[Scale acquisition in word spelling.] Retz.
Raven, J. (2000). The Raven's Progressive Matrices: Change and Stability over culture
and time. Cognitive Psychology, 41, 1-48. doi: doi:10.1006/cogp.1999.0735,
available online at http://www.idealibrary.com on
Raven, J. (Ed.). (1983). Standard Progressive Matrices. London: H.K. Lewis.
Rubin, H. (1988). Morphological knowledge and writing ability. Language and Speech,
31, 337-355.
Sangster, L., & Deacon, S. H. (2011). Development in Children's Sensibility on the Role
of Derivations in Spelling. Canadian Journal of Experimental Psychology, 65(2),
133-139.
Sénéchal, M. (2000). Morphological Effects in Children's Spelling of French Words.
Canadian Journal of Experimental Psychology, 54(2), 76-85.
Sénéchal, M., Basque, M., & Leclaire, T. (2006). Morphological knowledge as revealed
in children's spelling accuracy and reports of spelling strategies. Journal of
Experimental child psychology, 95, 231-254.
26
Seymour, P. (1997). Foundation of orthographic development,. In C. Perfetti, L. Rieben,
& M. Fayol (Eds.), Learning to spell: Research, theory and practice across
languages (pp. 319-337). Mahwah: Erlbaum Associates.
Seymour, P. (1999). Cognitive architecture of early readingp. 59-73. In I. Lundberg, F. E.
Tonnessen, & I. Austad (Eds.), Dyslexia, advances in theory and practice (pp. 59-
73). Dordrecht: Kluwer Academic Publisher.
Sprenger-Charolles, L., Siegel, L., Béchennec, D., & Serniclaes, W. (2003). Development
of phonological and orthographic processing in reading aloud, in silent reading
and in spelling: A four-year longitudinal study. Experimental Child Psychology,
84, 194-217.
Treiman, R., & Cassar, M. (1997). L'acquisition de l'orthographe en français. [The
acquisition of spelling in French] In L. Rieben, M. Fayol, & C. Perfetti (Eds.),
Des orthographes et leur acquisition [Of orthographies and their acquisition.]
(pp. 197-220). Lausanne, Paris: Delachaux et Niestlé.
Tyler, A., & Nagy, W. E. (1987). The acquisition of english derivational morphology.
Center for the Study of Reading: University of Illinois at Urbana-Champain.
Ziegler, J. C., & Goswami, U. (2005). Reading acquisition, developmental dyslexia and
skilled reading across languages: A psycholinguistic grain size theory.
Psycological Bulletin, 131(1), 3-29.
APPENDIX A
Words Graphemes in
prefixes Graphemes
in bases Morphogra
mmes Graphemes in
suffixes
immobile immobile
irréparable irréparable
arrondir arrondir
atterrir atterrir atterrir
alléger alléger
innombrable innombrable
déshabiller déshabiller déshabiller
illimitéillimité
embellir embellir embellir
décollage décollage
gagnant gagnant
gentillesse gentillesse gentillesse
banquier banquier banquier
chanceux chanceux chanceux
soirée soirée
violence violence
pâtisserie pâtisserie pâtisserie
affaiblissement affaiblissement affaiblissem
ent affaiblissement
regard regard
gout gout
bavard bavard
27
imprécis imprécis
début début
gratuit gratuit
retard retard
bord bord
talent talent
sourd sourd
livret livret
abandonner abandonner
encercler encercler
Dictation text
Un voyage de rêve
Au début*, le décollage de l’avion causa un léger vertige à Nicolas.
Cependant, il était excité d’être à bord de cet énorme appareil. Il ferma les
yeux et resta immobile pendant quelques minutes pour se calmer. Le
souvenir de l’instant où un mouvement imprécis de son adversaire violent
lui permit de conquérir la dernière planète, la Terre, et de devenir le plus
jeune gagnant de La conquête de la galaxie Morpho l’excita davantage. On
lui avait offert un billet gratuit pour le prochain championnat mondial. De
plus, son grand-père pâtissier lui avait même fait cette soirée-là un gâteau
embelli d’innombrables planètes au goût chocolaté. Il sourit et ouvrit les
yeux. L’hôtesse de l’air expliquait quelque chose dans un langage pour les
sourds. Ne comprenant rien, il jeta un regard à la personne qui était assise à
sa droite. Un homme, bavard comme un banquier, le salua
imperceptiblement. Nicolas lui sourit et détourna les yeux. À sa gauche, une
fillette déshabillait sa poupée pour la revêtir quelques minutes plus tard.
Plus loin, un garçonnet n’arrêtait pas d’encercler des lettres sur un livret.
Nicolas referma les yeux pour pouvoir penser aux stratégies que ses amis du
Cercle de l’Espace lui avaient suggérées. Ces stratégies et son talent
devraient alléger sa tâche de gagner. Il n’utiliserait jamais la violence, car il
n’était pas un faible. Il s’endormit en rêvant à ses figurines qui gagnaient
continuellement du terrain. Soudain, une panne irréparable de l’un de ses
vaisseaux permit à l’adversaire de faire le gain de plusieurs satellites. Il
n’avait plus de stratégies dans sa banque. Un terrible affaiblissement
l’envahit.
La main que l’hôtesse de l’air posa avec gentillesse sur son épaule le
réveilla. Fiou! Ce n’était qu’un rêve. Un plat couvert de pâtisseries
arrondies était déposé sur la tablette en face de lui. Il se sentit chanceux
d’être là et avait hâte à la compétition qui serait féroce. Ses chances de
28
gagner ne seraient pas illimitées, mais il poursuivrait sans abandonner.
Après quelques manœuvres du pilote, l’avion atterrit à New York. Nicolas,
habillé de son chandail du championnat descendit fièrement l’escalier de
l’avion.
*Words in bold were dictated.
29
