The Childrens Understanding of SVO Word Order - Essay Example

The Children’s Understanding of SVO Word Order Primarily, the title of the paper is very appropriate for its content. The authors made it appoint to make its readers be informed immediately of what was the study is all about by directly pointing to their target groups and on what aspect are the respondents were being tested and on what account they have focused their study. In short, the chosen title is suitable enough to briefly inform the reader of the topic of the research. There are two hypotheses that were tested in this paper. There were two groups that were compared by the authors in the paper. The first group is composed of French children who are 2 years in age and heard either high or low frequency verbs modelled in either SOV (subject-object-verb) or VSO (verb-subject-object) order. While the second group is likewise composed of French children who are 1 year older than the first group (3 years old) heard also either high or low frequency verbs modelled in either SOV or VSO order. The said word orders were both ungrammatical. The experiment raises two fundamental questions such as: 1) how do children perceive similarities between different lexical instantiations of related constructions and organize these into a network?; and 2) does a relative lack of similarity between related constructions affect the creation of a construction network? These said questions are addressed by the authors through replicating recent English weird word order production studies in French. To be specific, the authors intended to address five major questions: 1) Do French children show more robust preferences for grammatical word orders when using them with high frequency verbs? 2) Which versions of the French transitive construction do children use most often early on? 3) According to what features or similarities do French children organize transitive and related constructions? 4) Does a relative lack of similarity between soV and sVO transitives hinder the generalization of constructions in general and of the object role in transitives in particular? 5) Might differences in the input French and English children hear also affect the development of transitive constructions in these languages? Research method Primarily, the type of research being reviewed here is a causal research which appears to be the most appropriate to be done since the authors intend to find out the children’s understanding of SVO word order in French through hearing high or low frequency verbs. Specifically, the authors used the randomized controlled trial. Again, the authors chose the most appropriate method since they intend to inject an intervention and measure the effect of that intervention. Moreover, the research method employed has always been the best one since this is the most well-recognized method that could lead the researchers to an objective answer to the current research questions (Foster 2001, p.21). More specifically, the authors used weird word order methodology (Akhtar 1999) through using verbs of different frequencies, to determine whether children's use of word order as a grammatical marker depends upon the frequency of the lexical items being ordered. Specifically, the authors employed the similar process of the English weird word order study reported in Matthews et al. (2005) so as to achieve the same reliability and validity established by the said previous study. As previously mentioned, there were two verb frequency conditions: HIGH and LOW, two weird word orders: SOV and VSO and two age groups: TWO-YEAR-OLDS and THREE-YEAROLDS. This yields eight (2 ages x 2 verb frequencies x 2 word orders) between-subjects conditions. Each child was exposed to 4 verbs of either high or low frequency in either SOV or VSO order. Each verb had 12 corresponding video clips. Seven clips were described by E. For each of these clips E modelled the verb 4 times (giving a total of 28 weird models per verb). Five clips were used to elicit responses from the child. Thus, the child could respond 5 times for any given verb, giving a total of 20 possible responses per child. In addressing the first question, a weird word order study was conducted in French. For the second question, analysis made on the constructions children used when correcting the weird word orders to render an overview of which constructions French children prefer to use at the earliest stages of grammatical development, something which is currently little understood for French. In addressing the third question, authors compared the use of grammatical constructions for differences with age and according to the weird word order the children heard (SOV or VSO). To answer the fourth question, the authors have drawn cross-linguistic comparisons between the current study and the English weird word order study reported in Matthews et al. (2005) in terms of expression of the object. For the final question, a short analysis of French and English child-directed speech (CDS) is presented with the intention of elucidating how differences in the speech children hear might explain differences in their grammatical development. Clearly, the issue of problems in the validity of the study is not an issue in this case since the authors eliminated extraneous variables by testing all the children in a quiet room in their primary school or daycare centre in Lyon, France. Likewise, the research can be considered as highly valid and competent since the authors were able to defend that the total number of the respondents is already sufficient to warrant a general conclusion that their study is applicable to all French children whose age are similar to the used age range. Moreover, in a randomized controlled test, it is sufficient that the number and kinds of respondents are enough to draw conclusions from them. The respondents in this study consist of one hundred and twelve normally developing, monolingual French-speaking children (58 boys, 54 girls). The 56 children in the younger age condition were of a mean age of 2 years and 10 months (ranging from 2;3–3; 2). The 56 older children had a mean age of 3 years and 9 months (ranging from 3;3–4; 3). The authors ensured that all of the participants/respondents are within the age range. In this manner, the readers are further assured that the external variables would not have an issue in terms of the validity of the study. To show how meticulous the authors were in this study, supposedly, there should be 28 children more but they were not included in the study due to experimenter error or because they failed to complete the testing session or to produce any multiword utterances using any verb to describe the actions. As such, the method of selecting the participants or respondents had been very strict and complied well with the set standards. As mentioned before, all of these children were tested in a quiet room in their primary school or daycare centre in Lyon, France which makes the subject study reliable in terms of the possible presence of extraneous variables. As consistent with the aim to produce a reliable study, the authors assured the internal validity of the study. The following are the proof of high internal validity of the study: 1) Making standard to use eight transitive verbs to form two between-subjects conditions on the basis of verb frequency, specifically, four high frequency verbs and four low frequency verbs were identified using frequency counts from the LEXIQUE and BRULEX online lexical databases. The high frequency verbs were then identified in the French corpora available on the CHILDES database to check if two- and three-year-olds were likely to be familiar with them; 2) The verbs used were selected on the criteria that they should, as far as possible: (a) be exclusively transitive ; (b) have a semantic match in the corresponding frequency condition; (c) be able to take several animate subjects and objects; (d) map to equally complex actions;7 and (e) be equally nominalizable at all levels (with the verb being more frequent than any homonym nouns; 3) Making certain that each child saw exactly the same stimuli by using videos of hand puppets acting out the verbs. Since the verbs had been semantically matched across verb frequency conditions it was possible to use precisely the same video clips for both verb frequency conditions; 4) Each verb to be tested was enacted twelve times, each time by a different combination of hand puppets (a fox, a bear, a seal, a duck, an elephant and a giraffe). These enactments were compiled into a silent video that the experimenter and the child would take turns in describing. The enactments for each verb were compiled in pseudo-random order to guarantee that any enactment to be described by the child would not: (a) contain the seal (as this animal was too difficult to name for some children); or (b) contain the same agent or patient as the previous clip (this was essential to ensure that the child could not describe any enactment by simply repeating what the experimenter had said for the previous clip); 5) Counterbalancing the order of presentation of verbs such that, for each frequency condition, each verb was presented first, second, third or fourth an equal number of times for the experiment as a whole. All the videos were of precisely the same duration (13 minutes); 6) A transcription of the child’s utterances was made during the experiment by E and a second observer. These transcripts were checked against audio recordings of the experimental session; and 7) The authors employed a third investigator who is a native French speaker blind to the hypotheses of the experiment whereby he transcribed and coded 20% of the recordings. Reliability To further ensure the reliability of the study, in coding the transcripts it has been agreed that the first author will code all of the transcripts while a third investigator, a native French speaker blind to the hypotheses of the experiment, transcribed and coded 20% of the recordings. In such case, it can be said that the authors have protected the reliability of their research very well. Moreover, statistically, the reliability of the figures was very good as evidenced by the computation made through Cohen’s kappa revealing it 0.973 valid. In this sense, reliability of the data is not an issue in this research. Cohen's kappa coefficient is a method to statistically measure inter-rater agreement for qualitative (categorical) items. It believed to be a more robust measure than simple percent agreement calculation since Cohen’s kappa takes into account the agreement occurring by chance (Cohen & Jacob 1960 p.39). The way of analyzing the results in terms of mean proportions of responses is appropriate as it has been observed that some respondents are more talkative than others. Subsequently, the proportion of matches, single argument reversions and full reversions were analysed by ANOVA (with three separate 2 (age)x2 (frequency)x2 (order) ANOVAs (N=84)). Analysis of variance (ANOVA) is a stastistical method wherein a collection of models and their associated procedures, in which the observed variance is partitioned into components due to varying explanatory variables. This test is appropriate for the study since ANOVA gives a statistical test of whether the means of several groups are all equal, and therefore generalizes two-sample t-test to more than two groups (Ferguson and Takane 2005). The problem with this kind of test, however, is that statisticians would often make a different follow-up tests to verify the significant effect of the results of the variables measured. For this study, the test used can be referred to as valid because the conclusions cover the study population. The authors were able to clearly show this by making careful conclusions and by indicating to which the results would apply to. A valid experiment is one that fairly tests the hypothesis. In a valid experiment all variables are kept constant apart from those being investigated, all systematic errors have been eliminated and random errors are reduced by taking the mean of multiple measurements. An experiment could produce reliable results but be invalid. An unreliable experiment must be inaccurate, and invalid as a valid scientific experiment would produce reliable results in multiple trials (Butler, 2004). Moreover, the study can be considered as reliable and valid because the authors tried to use the same weird word order methodology of Akhtar (1999) wherein they used verbs of different frequencies and employed the similar process of the English weird word order study reported in Matthews et al. (2005). Results and data analysis In using ANOVA, it has been revealed that age have no significant interactions or effects. As for the matches, the ANOVA revealed a main effect of verb frequency. This repeats in French the effect of verb frequency on word order preferences found in English and is unfailing with the view that the ability to use word order as a grammatical marker surfaces out of lexically specific representations. There was no effect of modelled word order. For single argument reversions the ANOVA revealed a borderline effect of modelled word order only or there was no effect of verb frequency. On SVO reversions, ANOVA revealed a significant effect of verb frequency and a significant effect of modelled word order. To interpret the results, it has been concluded that: (1) children were more likely to adopt a weird word order if they heard lower frequency verbs, suggesting gradual learning; (2) children in the high frequency conditions tended to correct the ungrammatical model they heard to the closest grammatical alternative, suggesting different models activated different grammatical schemas; and (3) children were less likely to express the object of a transitive verb than were English children in an equivalent study, suggesting object expression is more difficult to master in French, perhaps because of its inconsistency in the input. These findings are discussed in the context of a usage-based model of language acquisition. On the aspect of ethical considerations, the research can be considered as made within the parameters of an ethical research as the authors remained honest and legal in all its procedures, maintained objectivity, integrity and carefullness with the research procedures especially that the respondents are all children and took steps to maintain competence in their field of research. This study can find its importance on the studies of psychology (particularly of social psychology) and sociology (particularly anthropology). It can be said that this study can be very helpful for further studies since the authors tried to discuss the results of their own experiment together with the results of the previous experiments and studies made before them that deal on different language, which is English. Though there are problems like having diifficulty for the authors to assess if the French children understood the low frequency verbs as verbs and that there was no effect of age was observed in the current study unlike other previous studies in English (Abbot-Smith et al., 2001; Akhtar, 1999; Matthews et al., 2005), the said potential criticisms are just of minor issues and it can also be argued that the research design was a causal one instead of a relational research and the latter would only be appropriate if it was part of the intention of the authors to study the relationship between the understanding of verbs and the age of the respondents. Over-all, it can be noted that the experiment is reliable and valid which makes the results highly competent. References Abbot-Smith, K, Lieven, EVM. & Tomasello, M 2001, “What preschool children do and do not do with ungrammatical word orders”, Cognitive Development , vol.16, pp.1–14. Akhtar, N 1999, “Acquiring basic word order : Evidence for data-driven learning of syntactic structure”, Journal of Child Language, vol.26, no.2, pp.339–56. Butler, M, 2004, “Optimizing Student Engagement and Results in the Quanta to Quarks Option”, http://science.uniserve.edu.au/school/curric/stage6/phys/stw2004/butler.pdf (Accessed 5 May 2009) Cohen, Jacob 1960, “A coefficient of agreement for nominal scales”, Educational and Psychological Measurement, vol.20, no.1, pp.37–46. Ferguson, G & Takane, Y, 2005, "Statistical Analysis in Psychology and Education", Sixth Edition. Montréal, Quebec: McGraw-Hill Ryerson Limited. Foster, C 2001, The ethics of medical research on humans, Cambridge University Press. Matthews, DE, Lieven, EVM, Theakston, A. & Tomasello, M 2005, “The role of frequency in the acquisition of English word order”, Cognitive Development, vol. 20, pp.121–136. Read More