Introduction: Design of the Studies
A quantitative design is an important component in measuring and rating the behaviors in pre-test and posttest experiments. Both articles under analysis have applied to quantitative research methods evaluate behavioral patterns and define the extent to which a specific invariable component can influence other variables.
In this respect, the studies refer to linguistics as the major factor influencing other behavioral patterns, as well as introducing shits to attitude and perception. The articles are based on experimental design because the participants of the study are involved into specifically designed situations for the purpose of defining whether these manipulations can change behavior.
However, the research instruments, along with the number of participants, independent and dependent variables, and sampling, differ significantly. At this point, the first article under consideration called Vocabulary Learning in an Automated Graded Reading Program applies to an cluster sampling because of the geographical advantages and it takes a long period of time (twelve weeks) to carry out the study.
In contrast, the second research entitled as Self-Identity Changes and English Learning among Chinese Undergraduates provides an overview of a stratified sampling that reveals different patterns of behavior with regard to such variables as sex, major course, and age, which are not presented in the first study.
In order to evaluate the relevancy and validity of the methodology, it is purposeful to define the correlation between the research design and the results of the study.
Methodology Overview
Brief overview of research methods
In the study by Huang (2007), the primary emphasis has been placed on the selection of the articles that would fit the requirements of the research. In order to choose the 16 articles, such factors as students’ background and word frequency have been used. Introducing such computerized device as Textgrader, on the one hand, is an effective approach to define the size, the number of new words to be introduced, as well as other parameters related to the topic of the discussion.
On the other hand, use of automated analysis does not contribute to a consistent evaluation of human factors, since students can have different levels of word exposure and knowledge background to memorize words and infer them from the context.
The problem is especially critical as far as the sampling population is concerned. The software involved into the study, however, cannot be considered cost-effective because it requires installation of specific computerized devices.
Regarding the methodological instruments, there is a strong focus made on searching for the texts for students to read. However, lack of tools necessary to select a sample introduces ambiguity.
In other words, the study does not provide a sufficient explanation why the fixed sample has been chosen to take part into word lists reading program. In additional, the author has failed to evaluate the overall background of the participants and split those into specific groups with regard to the identified chosen word lists.
In general, the research is a longitudinal study because it is premised on repetitive actions focused on a particular group of students to measure any changes and deviations from the established norms.
Introducing experimental program provides more perspectives for analysis in terms of introduced variables and the sample population. Pre-test and posttest stages have also been introduced to define any improvements among the students.
Presence of questionnaires providing an extensive overview of the participants has contributed to the analysis of the learners’ attitude to the Internet use in learning English. Though two questionnaires have been designed to estimate the learners’ attitude toward reading, as well as background, no changes have been made to the overall design of the study.
At this point, all the participants have relatively different levels of computer access at home and different experiences in reading English texts. Therefore, the study does not demonstrate how this information could be applied to fit the experimental design of the study.
In contrast to the first study, Yihong et al. (2005) have introduced a stratified population sampling enabling the researchers to draw conclusions about particular subgroups that cannot be defined in a generalized random sample. The chosen sample approach contributes to accumulating data across specified groups.
Because the sample has been chosen from 30 universities, it is possible to infer that the sample is the closest to the general population criteria. Finally, the choice of a stratum enables choosing various sampling approaches, including the most cost-effective ones.
In this respect, the researchers have applied mostly to questionnaires to include 24 statements about different categories of change based on existing bilingual theory. As a result of the quantitative methods used in the form of questionnaires and variance analysis, the research has managed to achieve the highest results.
Using questionnaire as a methodological instrument has created a number of benefits for the researchers in terms of information availability and statistical analysis.
Though there was lack of computerized tools, the research is still valid and reliable because of the variety of questions introduced into the survey. A set of observations gained from the questionnaire has become the major source of quantitative analysis. It is also important to note that the researchers have managed to present sufficient information for statistical evaluation due to the categories introduced into the study.
Judging from the choice of the sample and methods of research, the article refers to linguistic sociological study with its aim to measure the identities of different participants within a specific sample. Nevertheless, the author recognizes that using standard deviations test to define how these categories can be applied to evaluate the identity changes in correlation with the other independent variables.
Data Collection Techniques
Regarding the quantitative methodologies applied in the first study, it should be noted that many frameworks implemented are relied on other experimental studies related to linguistics and learning techniques. At this point, the overview of related experiments and observational studies has greatly contributed to the exposition of quantitative data in the given study.
As per the second study, the researchers have applied to conventional methodologies to analyze the collected data and to choose the sample population. Overall, both articles have presented an extensive overview of relevant information collection methods.
However, the second research initiated by Yihong et al. (2005) represents data with greater precision. Specifically, the availability of different charts and tables creates a wider insight into the accuracy of the data introduced, as well as stronger understanding of how different dependent and independent variables are interconnected.
While evaluating the data collection hypotheses in more detail, it should be stressed that Huang (2007) has managed to integrate clearly stated purposes of the research and test those questions effectively. In such manner, the scholar has exposed a clear frame of the study. The second article has also introduced a clear thesis statement, but there is no specific identification of question to be tested.
Despite the drawback, both articles have clearly introduced the data collection instruments, as well as data collection procedures.
Hence, the first article applies to computerized tools to evaluate the appropriateness of the tools, whereas the second study makes use of the previous researches and theories to make up a questionnaire. Regardless of the differences in the methodological instruments, they still add value to the research due to the rates and scores provided in the study evaluation.
The data collection procedures fully conform to the purpose of the research in terms of normality of distribution, homogeneity of variance, and levels of measurements.
However, the first article fails to reveal the powerful data analysis techniques with regard to statistical procedure because no specific graphs and tables have been presented. As per the second study, the data analysis criteria have been full met with regard to the availability of statistical procedures.
Because the two researchers are premised on different approaches to data recording, the degree of errors differs as well.
At this point, the first article is mostly based on computerized data assessment, which provides lower probability of error occurrence whereas the second article is more based on manual data recording which might involve more errors because of the human factor. Therefore, the second article provides lower level of reliability.
Conclusion: Evaluation Of Methodologies
While analyzing the above-presented methodological instruments for data collection and analysis, it can be concluded that both articles have suitable quantitative research designs because they fully meet the requirements of the research questions. More importantly, all the data collection procedures correspond to accepted procedures of experimental design.
As per the sample population, the research articles outline different approaches to choosing the sample due to the various objectives of the study. In particular, the first article represents cluster sampling to carry out a longitudinal study, whereas the second article introduces a stratified sampling that perfectly suits a linguistic sociological nature of the research.
Data collection procedures and instruments are powerful enough to measure the identified attitudinal patterns and important characteristics with adequate accuracy. Hence, using computerized tools in the first study enables the research to identify the texts that would disclose the major characteristics of text required for the program.
The statements identified in the second study also offer a full picture of behavioral and changing patterns influencing students who study English. The difficulties, however, might occur because no evidence and arguments have been integrated with regard to the number of the participants chosen in the first study.
Both articles add value to other existing researches in the sphere of linguistic and contribute significantly the frameworks needed for learning languages. Specifically, because the articles are focused on the patterns and measurements of scores and rates, they can be compared with existing statistical data referred to similar topics.
Also, both studies manage to accurately identify the actual purposes of the instruments used, as well as find a close correlation between existing dependent and independent variables. What is more important is that separate findings from the studies can be used to be elaborate those in other researches.
References
Huang, H.-T. (2007). Vocabulary Learning in an Automated Graded Reading Program. Language Learning & Technology. 11(3), 64-82.
Yihong, G., Ying, C., Yuan, Z., & Yan, Z. (2005). Self-Identity Changes and English Learning among Chinese Undergraduates. World Englishes. 24(1), 39-51.