The following is an analysis of the method of the follwoign study: Holt et al (2012) investigated the epidemiological characteristics of a group of children who were hard of hearing. They wanted to identify the predictor variables that determined timely follow-up after a failed newborn hearing screening, and variables that hindered timely follow-up. The authors studied 193 children from three states each of whom had hearing loss and did not pass the newborn hearing screening. Available records were used to capture ages of confirmation of hearing loss, hearing aid fitting, and entry into early intervention. Linear regression models were used to investigate relationships among individual predictor variables and age at each follow-up benchmark.
¶ … population included in the study, the methodology employed by the investigators, the data analysis in the study, the authors' interpretations of their results, limitations of the study, etc. Finally how well or how poorly the study was performed.
Holt et al. (2012) investigated the epidemiological characteristics of a group of children who were hard of hearing. They wanted to identify the predictor variables that determined timely follow-up after a failed newborn hearing screening, and variables that hindered timely follow-up.
The authors studied 193 children from three states each of whom had hearing loss and did not pass the newborn hearing screening. Available records were used to capture ages of confirmation of hearing loss, hearing aid fitting, and entry into early intervention. Linear regression models were used to investigate relationships among individual predictor variables and age at each follow-up benchmark.
The authors discovered that of all variables only level of mother's higher education was significantly related to timely follow-up and fitting of hearing aids. Severity of hearing loss was not. There were no particular variables that correlated with age of entry that children were tested. Each recommended benchmark was met by a majority of children, but only one third of the studied population met all of the benchmarks within the recommended time frame.
The authors recommended that specific attention be focused on children of underprivileged communities need extra support in navigating steps that follow failed newborn hearing screening.
Analysis of Study
The study was done on an appreciably large population (n- 193). This gave it a certain amount of reliability and the possibly of concluding statistical significance. A great many variables too were taken into consideration in order to exclude all possible bias and possibility of outcome happening by chance. Batteries that were used to test each of these considerations included: (a) family and community factors (e.g., SES, race, ethnicity, service access, parental education); (b) child factors (e.g., gender, severity, and type of hearing loss; etiology); (c) child outcomes
(e.g., receptive and expressive language, speech perception and production, psychosocial development, academic abilities); and (d) intervention characteristics (e.g., audio logical, therapeutic, and educational). Test batteries were also developmental-appropriate and these included: normative-based tests, speech and language elicitation tasks, language sampling, and parent and service provider questionnaires. Research was thorough to the extent that at each visit, children completed a comprehensive pediatric audiological evaluation.
Furthermore, population was also selected from three different states which further excluded possibility of chance and diversified the population. The population origin too was not a convenience sample (i.e. from a ready-at-hand source) but was rather amassed from a huge pool extracted from eight states:
State EHDI coordinators, audiologists, early intervention specialists, and educators assisted with recruiting HH children. Children with normal hearing (NH) were recruited from databases of past research participants, fliers in community centers and child care centers, advertisements in newsletters, and word of mouth. (165)
Conclusions criteria were exhaustive: (a) permanent bilateral hearing loss of any type (sensorineural, mixed, conductive); (b) better ear pure-tone average (PTA) (500 Hz, 1000 Hz, 2000 Hz, and 4000 Hz) between 25 dB HL and 75 dB HL; (c) entry ages between 0;6 and 6;11; (d) no known significant sensory or developmental disorders; and (e) at least one primary caregiver who speaks English in the home.
The statistical method too was appropriate. Linear regression analyses whether a positive or negative significance between the two variables is and an accelerated longitudinal design was used in order to maximize the amount of developmental and cross-sectional data that could be collected in a relatively short period of time. IRB approval was sought to prevent possible ethical concerns. Finally, data was collected from multiple sources: "The researchers collected retrospective historical, medical, audiological, and education data to supplement the prospective data." (165). this short-circuited the possibility of bias and error of data being collected merely from one or fewer sources.
The key problem that I see with the study is that the eight states that the population was extracted from possess a homogenous population with similar characteristics. They do not provide the diversity that is needed to make generalities from this to similar studies. This fact can be evidenced by the demographics shown in Table 1 which indicate that the Caucasian / White population is significantly disproportionate to that of any minority race (White= 151; Black is next with n=12). The authors' conclusions, therefore, of timeliness being correlated to education may be incorrect. It may be that race / ethnicity may be the determining factor, but this aspect was not looked into.
You’re 87% through this paper. Sign up to read the full paper.
Sign Up Now — Instant Access Already a member? Log inAlways verify citation format against your institution’s current style guide requirements.