Hart, B. And Risely, T

Hart, B. And Risely, T (2003). The early catastrophe: The 30 million word gap by age 3. American Educator, Spring

Fairly commenting on an investigator's research endeavor is a task that must be taken seriously. Although it is quite easy to have an opinion of another's research, it is something quite different to be able to evaluate the research activity in terms of topic specificity and soundness, intent or purpose, data analysis, and informational importance. The focus of this paper was on whether or not the research investigators of the above cited research publication were prudent in stating a research question and a testable hypothesis along with informing the reader of the chosen research design, statistical data analysis and reporting the results, limitations, limitations and implications for future practice - all of which must lead to a best fit research decision.

The authors of this particular research report not only failed to state a research question and testable null hypothesis but selected a sample (N=42 families) on a non-random basis. In fact the sample selection was reported as being "pre-selected." As such any results garnered from a statistical data analysis can only be inferred back to the selected population and not to a wider universe of language growth deficient children. In fact, the authors set out to examine language deficiency of lower income children yet, included in their analysis a disproportionate number of upper income (13), middle income (10), lower income families (13), and welfare (6). Not only was there disparity among family selection the authors failed to report how many children were included in each of the four socio-economic status categories, thus producing error contamination of the results.

In addition to the errors associated with failure to state a research question and testable null hypothesis the authors of the article failed to alert the readers that a cross-sectional research investigation is point in time sensitive and as such greater population inferences are not possible. Making inferences to a greater population can only take place when the research investigator(s) conduct a longitudinal or experimental study that is sample selection appropriate and statistical processes are used to analyze the data. In fact, the authors began the article by stating that the research investigation was cross-sectional and later termed the research investigation a longitudinal study. Along with the erred research design and design confusion were the authors' lack of appropriate data selection procedures (i.e., non-random selection) and lack of measurement accurateness. The authors reported that their data came from observations over a six-year period of time and the data was then analyzed. However, the authors failed to inform the reader how these observations were quantified with respect to numeric values garnered from the observations. According to the authors correlations were run to determine certain relationships; however, the sought after relationships were never defined or adequately reported upon. Such an omission renders the study without sufficient merit.