Longitudinal Receptive American Sign Language Skills Across a Diverse Deaf Student Body

Abstract

This article presents results of a longitudinal study of receptive American Sign Language (ASL) skills for a large portion of the student body at a residential school for the deaf across four consecutive years. Scores were analyzed by age, gender, parental hearing status, years attending the residential school, and presence of a disability (i.e., deaf with a disability). Years 1 through 4 included the ASL Receptive Skills Test (ASL-RST); Years 2 through 4 also included the Receptive Test of ASL (RT-ASL). Student performance for both measures positively correlated with age; deaf students with deaf parents scored higher than their same-age peers with hearing parents in some instances but not others; and those with a documented disability tended to score lower than their peers without disabilities. These results provide longitudinal findings across a diverse segment of the deaf/hard of hearing residential school population.

Recent legislation in education, including the No Child Left Behind Act (2002), Individuals with Disabilities Education Improvement Act (IDEIA, 2004), and the Common Core State Standards Initiative (2010), as well as the new professional accreditation organization for teacher education programs in the United States (Council for the Accreditation of Educator Preparation, 2010) and the national teacher preparation assessment (edTPA; Stanford Center for Assessment, Learning, and Equity, 2013) call for data-driven instructional decisions and evidence-based instructional practices that are aligned with students’ current skill levels. Taken together, particularly for deaf education, these measures call for educators who can assess students’ language skills and adjust their language use accordingly during communication and instruction for effective academic outcomes (Haug, 2005). Teachers and researchers in deaf education have acknowledged a lack of available and efficient assessments of deaf students’ sign language skills, which limits data-based sign language instruction (Herman, 1998; Maller, Singleton, Supalla, & Wix, 1999; Mann & Prinz, 2006; Mann, Roy, & Marshall, 2013; Singleton & Supalla, 2011).

Documentation of students’ sign language skill levels is challenging for a number of reasons. Few assessments exist that are available and efficient to administer and score (Haug, 2005; Haug & Mann, 2005; Singleton & Supalla, 2011), and until recently most studies investigated the sign language skills of deaf children with deaf parents (DODP), which excludes the vast majority of deaf students (Mitchell & Karchmer, 2005) who tend to acquire American Sign Language (ASL) at later ages (Mayberry & Eichen, 1991; Musselman & Tane Akamatsu, 1999), and who continue to develop ASL skills long after their peers with deaf parents (Beal-Alvarez, 2014). Based on responses to the prompt communication mode primarily used to teach [D/HH] student, Gallaudet Research Institute’s (GRI) Annual Survey of Deaf and Hard of Hearing Children and Youth (2009, 2011, 2013) reported that a large portion of deaf students use sign language for instruction, either alone or paired with spoken language (i.e., simultaneous communication), but the actual estimate varies across sample size and year. In 2009, about 46% of 36,700 students used sign language; in 2011, about 40% of 37,800 did so; and in 2013, about 28%. GRI’s survey is estimated to represent about 65% of deaf/hard of hearing children in the United States (Knoors & Marschark, 2012), although this likely varies by total number sampled each year. However, educators have limited data on what a child should be able to comprehend and express across specific ages (see Maller et al., 1999, for a review of children’s ASL production).

Children who communicate via sign language frequently vary in their comprehension skills of that language (Allen & Enns, 2013; Enns, Hall, Isaac, & MacDonald, 2007; Prinz & Strong, 1998) due to factors such as lack of sign language at home (Moeller & Luetke-Stahlman, 1990; Moeller & Schick, 2006), age when they started learning ASL (Enns et al., 2007; Mayberry, 1993; Mayberry & Eichen, 1991; Mayberry & Lock, 2003), exposure to fluent ASL models in school (Schick, Williams, & Kupermintz, 2006), and their cognitive development and maturation processes (Berent, 1988). This variation in language experiences creates a population of children more varied in their language development (i.e., ASL) and comprehension than their typically hearing peers, who have been exposed to fully accessible spoken language from birth (Maller et al., 1999; Mann et al., 2013; Marshall, Rowley, Mason, Herman, & Morgan, 2013). Additionally, when asked, most teachers reported using expressive sign language assessments with students, including video recordings and observation checklists, but no receptive measures (Mann & Prinz, 2006). They were aware of the need for assessments of sign language to drive instruction; yet they felt linguistically inept at ASL assessment (Mann & Prinz, 2006). Researchers (Allen & Enns, 2013; Maller et al., 1999) have called for efficient receptive ASL measures as one part of documenting students’ ASL skills when they enter an educational program that utilizes an ASL approach, including decisions related to educational placement, progress monitoring, and accurate reporting of children’s language development (Allen & Enns, 2013).

Researchers also have called for a redefinition of assessment “norms” for ASL assessments, given the small size of and variation within the deaf student population (Hermans, Knoors, & Verhoeven, 2010; Mann & Haug, 2015; Mann et al., 2013; Singleton & Supalla, 2011), and questioned whether norms for every subgroup of deaf students can be developed (Hermans et al., 2010). Mann and Haug (2015) noted that the small size of the deaf population “poses a number of limitations for test developers when it comes to applying common statistical procedures to establish psychometric properties of a text to assure its reliability and validity” (p. 484). Mann and colleagues (2013) stated “it may be necessary to consider treating the variable signing experiences seen in the majority of deaf language users as normative” (pp. 94–95). Researchers have called for the investigation of students’ ASL skills related to gender, parental hearing status, and disabilities (Hermans et al., 2010; Johnson, 2004; Mann et al., 2013), with the suggestion of longitudinal “profiles” of students to examine the effects of these factors on the development of ASL skills over time (Allen & Enns, 2013; Beal-Alvarez, 2014; Mann et al., 2013). The purpose of the present study was to investigate longitudinal changes in students’ receptive ASL skills across subgroups within a convenience sample of a diverse student body at a residential school for the deaf. Below, I review the available literature on students’ receptive ASL development and describe the procedures of the present study.

Receptive ASL Assessment and Outcomes

Recently, researchers noted the “strong psychometric properties” (Allen & Enns, 2013, p. 66) of the British Sign Language (BSL) Receptive Skills Test (BSL-RST; Herman, Holmes, & Woll, 1999) and adapted it for use with ASL signers (ASL Receptive Skills Test, or ASL-RST; Enns, Zimmer, Boudreault, Rabu, & Broszeit, 2013). The standardization sample for the BSL-RST included 135 children, 3–11 years of age, from England, Scotland, and Ireland, the majority of whom were native signers, with the remaining students exposed to BSL prior to 5 years of age (Herman & Roy, 2006). Additionally, Herman and colleagues analyzed a second BSL-RST data set from 162 children, 3–14 years, from England and Wales who were tested by 18 different educational professionals. The majority had hearing parents. For both groups, scores increased with age, although there was greater variability and overall generally lower scores in the second sample. Those with deaf parents outscored their peers with hearing parents in both samples and girls outperformed boys in both samples, although significance emerged only for the second sample. Herman and Roy reported concurrent validity via a strong positive correlation between BSL-RST scores and Edinburgh Reading Test scores for 11 children (r = .70, p = .02). They reported construct validity via a significant relation between test administrator ratings of 19 children’s BSL comprehension based on their experiences with those children and their BSL-RST scores from Sample 2.

Enns et al.’s (2013) adapted ASL-RST is intended for students 3–13 years of age and measures ASL receptive skills in eight grammatical categories: (a) number/distribution (e.g., TWO ROWS (BEDS)); (b) negation (e.g., (NO) SLEEP and NOT-YET HAT); (c) noun-verb distinction (e.g., DRIVING and CHAIR); (d) spatial verbs-location (e.g., TABLE BALL ON) and spatial verbs-action (e.g., TWO-PEOPLE-MEET); (e) size and shape classifiers (e.g., THIN-STRIPES-DOWN-SHIRT); (f) handle classifiers (e.g., HOLD-UMBRELLA-WALKING); (g) role shift (e.g., TAP-GIRL, GIRL-TURN-LOOK); and (h) conditionals (e.g., IF RAIN, GAME CANCEL). First, participants identify 20 pictures to ensure they are familiar with stimuli in the assessment items; then they watch a sequence of 42 signed video clips on a computer (approximately 3s each) that present phrases in ASL and point to one of four pictures on the computer screen that corresponds with the signed phrase.

Enns and colleagues (2013) developed standard scores for the ASL-RST from a sample of 203 students, native (n = 77) and near-native signers (i.e., exposed to ASL by 3 years of age; n = 126) with deaf and/or hearing parents and no diagnosed disabilities, 3–13 years of age. They reported that score strongly correlated with age (r = .82) and marginal maximum likelihood reliability correlation of r = .88 (standard deviation [SD] = 15.15). Allen and Enns (2013) assessed 160 preschool children, 3–5 years of age, 41% who used sign language only and 50% of whom used speech and sign at home, from across 23 states. Half of the children had hearing parents and half had at least one deaf parent. Allen and Enns reported that children whose parents used sign at home (based on parental response), regardless of parental hearing status, performed significantly better than those who did not use sign language at home. Allen and Enns reported strong internal consistency among test items (Cronbach’s alpha = .96).

Scores across grammatical categories were not reported for the standardization sample or the younger sample, although Allen and Enns (2013) reported Cronbach’s alpha ranged from .15 to .86 across categories. They also reported steeper learning slopes and greater levels of mastery for number-distribution, negation, and SASSes compared to role shift, handle classifiers, and conditionals across the younger sample, suggesting negatives and number/distribution are acquired earlier (Hoffmeister, 1978; Meier, 1982) and more complex structures such as conditionals and role shift are developed later (Emmorey, 2000; Morgan, 2002; Reilly, McIntire, & Bellugi, 1990; Schick, 2010). Allen and Enns suggested that children “grow” into the ASL-RST between the ages of 3 and 5. However, the usefulness of this assessment with children beyond 11–12 years of age may be limited unless children outside the age range are suspected to have language delays (Beal-Alvarez, 2014; Enns et al., 2013).

Beal-Alvarez (2014) investigated ASL-RST (Enns et al., 2013) performance across a residential student body population of DODP and DOHP students, aged 6–22 years, and reported that scores strongly correlated with age for younger (i.e., 6–13 years) DOHP (r = .79) and DODP (r = .76), but correlations were not significant for older DOHP students (i.e., 14–22 years; r = .16). Students within the 6- to 13-year standard score range scored within 1 SD (i.e., ±1 SD below or above the mean; Enns et al., 2013) of their age-related standard scores. However, even the oldest students did not reach ceiling on an assessment designed for students up to 13 years of age; on average they scored around 75%. Across the grammatical categories, correlations were significant only for the younger students and few trends in category performance were evident for the cross-sectional data, although variation in performance decreased with age. Beal-Alvarez (2014) noted that based on an error analysis, over half of the students were incorrect on the last 12 items, which are deemed the most difficult (Enns et al., 2013) and are spread across six grammatical categories. No scores of students who were deaf with disabilities (DWD) were included in the published data.

Using nine receptive and expressive subtests of the Sign Language of the Netherlands (SLN) test, Hermans et al. (2010) reported that DODP (n = 32) significantly outperformed DOHP (n = 298) and there was a strong correlation between age and scores and that girls (n = 156) outperformed boys (n = 174) on all nine tests, similar to previous findings for measures of BSL (163 girls and 153 boys; Herman & Roy, 2006) and spoken language with deaf (Easterbrooks & O’Rourke, 2001) and typically hearing children (Reilly et al., 2010). However, Maller and colleagues (1999) reported no gender effects for the expressive ASL Proficiency Assessment (ASL-PA) and Haug (2011) reported no significant gender effects with the DGS (German Sign Language) Receptive Skills Test. Although Hermans et al. (2010) reported that 163 children were tested with the nine subtests three consecutive years and 76 children were tested for two consecutive years, no correlations were reported for longitudinal SLN scores. However, Ormel (2008) reported that students’ longitudinal receptive SLN scores correlated with students’ reading comprehension 1 and 2 years later, similar to previous findings of relations between ASL and reading skills (Easterbrooks & Huston, 2008; Freel et al., 2011; Hoffmeister, de Villiers, Engen, & Topol, 1998; Padden & Ramsey, 1998; Strong & Prinz, 1997).

Presently, only one longitudinal study of deaf children’s language development appears available: the Longitudinal Outcomes of Children with Hearing Impairment (LOCHI) study in Australia, which aims to capture language data from 451 children at 3, 5, 9, and 12 years of age and examine language factors related to demographic variables (Ching & Dillon, 2013). Children in the study predominantly used spoken language, although one quarter used it with sign support in the home and/or early education setting. Nearly all children used hearing aids or cochlear implants (CIs). The assessments measured language skills that rely on auditory input to develop and were administered in the child’s mode of communication. Ching and Dillon calculated a global language score created from nine measures, including speech production, receptive and expressive language (Peabody Picture Vocabulary Test [PPVT]; Preschool Language Scale Fourth Edition [PLS-4]), and psychosocial development, with a mean score of 100 and SD of 15. The mean score across the cohort of 3-year-olds was 74.6 with a SD of 17.1. These initial scores document a language gap of more than 1 SD between young deaf children who used spoken language and their typically hearing peers.

Deaf With Disabilities

To date, few studies have investigated the receptive sign language skills of deaf students with disabilities (DWD). Across available studies, the incidence of a disability co-occurring with a hearing loss is between 20% and 40% (Berrettini et al., 2008; Cupples et al., 2014; GRI, 2013; Kennedy et al., 2006; Picard, 2004). Certain disabilities appear more prevalent than others. A sample of 23,731 deaf children, 40% of whom had a disability, included the following prevalence of diagnoses: intellectual disability (ID; 8.8%), specific learning disability (SLD; 7.2%), other health impairment (OHI; 7.1%), “other condition” (6.9%), developmental delay (DD; 6%), attention disorder (ADHD/ADD; 5.4%), visual impairment (5%), speech/language impairment (SLI; 3.3%), autism spectrum disorder (ASD; 2.2%), emotional disturbance (2.1%), and deafblind (1.2%; GRI, 2013). These percentages remained consistent across GRI samples of more than 36,000 students from 2008 and 2011. These data may overrepresent deaf students with disabilities, as they include more students from residential schools (Knoors & Marschark, 2012), which frequently have more DWD students than local public schools.

Nearly all available studies of the language skills of DWD focus on spoken language skills of children who received CIs (see Cupples et al., 2014, for a review). Based on limited studies, DWD with CIs appear to progress more slowly and with greater variation in their auditory-linguistic skills than their deaf peers without disabilities (Hamzavi, Baumgartner, Pok, Franz, & Gstoettner, 2003; Kaga, Shindo, Tamai, & Tanaka, 2007; Waltzman, Scalchunes, & Cohen, 2000). Cupples and colleagues (2014) reported on the language scores of 119 three-year-old deaf children with disabilities from the LOCHI study, which represented 23% of the children in the study. They divided the children into two categories of disabilities: those with autism, cerebral palsy (CP), or DD paired with another syndrome or condition; and those with DD (without another syndrome/condition), vision loss, speech impairment, or various syndromes not entailing DD. Three quarters of the children used hearing aids and one quarter used CIs. A larger portion of DWD used sign-supported speech (43%) than their peers without disabilities in the overall LOCHI study. All measures were identical. Cupples and colleagues found that children with autism, CP, or DD were less likely to complete the PPVT and achieved lower receptive and expressive scores than the children with other disabilities with no difference by gender.

Fewer results exist regarding DWD’s sign language skills. Mann and colleagues (2013) noted that “little is known about the language development of deaf children with additional disabilities, specifically how any of these disabilities affects vocabulary development over and beyond the effect of their primary hearing deficit” (p. 93). They addressed this by administering two receptive and one expressive web-based BSL vocabulary measures they developed to deaf students 4–17 years of age, both without (n = 25) and with (n = 18) 12 various disabilities. Assessments were administered by students’ teachers of the deaf or speech language pathologists who received assessment training. Mann et al. reported no significant differences in scores among deaf and DWD participants across measures and a strong correlation between age and score. When compared to a pilot sample of “strong” signers (i.e., native-like, N = 24), deaf and DWD groups had similar means as the strong signers but greater variation in scores. Although Mann and colleagues acknowledge that disabilities have an effect on vocabulary acquisition, they stated that the lack of a significant effect for having a disability in their study “suggests that for deaf children as a whole this particular factor [having a disability] is not as important for vocabulary acquisition as other factors might be, specifically the impact of their primary deficit of hearing loss” (p. 111).

Receptive abilities, or understanding the language, develop prior to productive abilities, or expressing the language. Limited results are available for age of production of ASL structures (see Baker, van den Bogaerde, & Woll, 2008; Chen-Pichler, 2012, for reviews). Maller et al. (1999) reviewed 17 acquisition studies related to DODP’s production of eight ASL morpho-syntactic linguistic structures and concluded that children acquired/produced aspects within the ASL-RST as follows: aspect and number (i.e., number distribution) 3;7–4;8, although they “can still exhibit some grammatical errors past age 5” (Maller et al., 1999, p. 255); noun-verb pairs with 70% accuracy at 3;0–3;11; verbs of motion at 3–12 years of age (see Beal-Alvarez, 2014, for an overview), which have longer trajectories of development based on complexity; referential shift is mastered around 4;4 (Loew, 1980), although its use within certain functions may require more time (see Quinto-Pozos, Forber-Pratt, & Singleton, 2011); and use of nonmanual markers (including conditionals) around 3;11, with additional time needed to achieve mastery. To investigate aspects of children’s ASL acquisition, Maller and colleagues used three expressive tasks (interview with signing adult, peer interaction, and story retelling of a cartoon) with 6- to 12-year-old native (n = 28), nonnative (n = 37), and manually coded English signers (n = 15). They reported significant differences in mean and SD scores across groups: native ASL signers scored highest (M = 17.50, SD = 1.93), followed by nonnative ASL signers (M = 15.03, SD = 3.20) and MCE users (M = 8.00, SD = 2.30), and less variation in native signers than nonnative signers. They reported no significant gender or age differences across groups for these expressive tasks, unlike Hermans et al. (2010) and Herman and Roy (2006) for receptive tasks.

Using a BSL expressive semantic fluency task (i.e., generation of animals and foods), Marshall et al. (2013) reported that 13 deaf children, aged 4–15 years and diagnosed with SLI (defined as “children who are not acquiring sign language as well as would be expected in comparison to peers who have had the same (delayed) language experience,” p. 197) performed similarly to their peers without SLI but the former appeared to make word-retrieval errors and accessed signs less efficiently. In contrast to Mann and colleagues’ (2013) receptive findings for DWD students, Marshall et al.’s results suggest differences in expressive performance between deaf and DWD students.

Longitudinal development

Longitudinal studies of students’ ASL development over time are not readily available in the published literature; however, Lange, Lane-Outlaw, Lange, and Sherwood (2013) investigated longitudinal academic growth in deaf students’ reading and math scores across 4 years for students at a pre-k-21 charter school for the deaf that used a bilingual ASL/English model of instruction. They reported no difference in reading or math growth by gender or parental hearing status but significantly less growth in both areas for DWD (i.e., visual impairment, DD, and behavior disorder). Finally, they reported that academic growth for deaf students was initially slower than their typically hearing peers, but that deaf students eventually exceeded comparison group scores (after 8.2 years for reading and 2.5 years for math) and attributed this lag to the time needed to acquire academic proficiency in a second language (i.e., English or ASL for these students).

No results are currently available to guide instructional decisions regarding students’ ASL skills and how they should progress across ages. Available studies are limited by small sample sizes and lack of generalizability across the deaf student population. The current population of deaf students who use sign language is diverse; how does this diversity affect their comprehension of ASL across time and therefore educators’ use of ASL during instruction? The ASL comprehension of a deaf child with deaf parents who signed to her from birth will look much different than that of a child who began with spoken language and transitioned to the use of ASL as a teenager. The present study aims to document the receptive ASL skills of a residential school population across 4 years. Beal-Alvarez (2014) presented an initial snapshot of students’ receptive ASL skills; the present study analyzes the longitudinal language development of these students one, two, and years later, including DWD students whose results were not previously presented. My research questions were: (a) How are deaf students’ receptive ASL skills (i.e., performance on the ASL-RST and the Receptive Test of ASL [RT-ASL]) affected by age, gender, parental hearing status, and disabilities? and (b) How do students’ receptive ASL skills change across four academic years?

Methods

Setting and Participants

All participants were school-aged students at a residential school for the deaf in the southeastern United States. The student body consisted of around 110 students per year, pre-kindergarten through high school, from more than 150 different counties within the state; however, on average about 15% of the student population changed every year. Students stayed in dorms during the week and returned home each weekend (with the exception of 10 day-students who participated in the study). Students were allowed to continue at the school through their 22nd birthday. Teachers at the school used ASL for instruction and were required to have at least an intermediate score on the Sign Language Proficiency Interview (SLPI; Newell, Caccamise, Boardman, & Holcomb, 1983). Thirty percent of the teaching staff was deaf. Students attended ASL class as a 50-min elective course a few times a week.

Inclusionary criteria for the study included only that students attended school at the research site and that they were able to maintain attention throughout the tasks. Table 1 shows the number of students included each year and longitudinally across 4 years. The most frequent reasons for the fluctuation in attendance were graduation and changing schools; a small number (3–4 students) were absent from school during assessments each year. Twelve students with disabilities were included. The majority were diagnosed with mild (MID; n = 5) and moderate intellectual disabilities (MOD; n = 4). Two were diagnosed with autism (AUT) and one with Waardenburg syndrome, classified under OHI. At the time of initial assessment (Year 1), students had attended the residential school for a range of a few months to all of their educational years and their ages ranged from 6;4 (years; months) to 22;2. Data on hearing level were not collected, as degree of hearing loss is not an important factor in studies of sign language syntax (Baker et al., 2008), and data on age of acquisition of sign language were not available.

Table 1.

Number of students assessed across each year and across all 4 years

	Years
	1	2	3	4	1, 2, 3, and 4
DOHP	75	69	76	71	30
DODP	9	6	2	5	2
DWD	11	12	11	11	7
Total	95	87	89	87	39

Note. DODP = deaf of deaf parents; DOHP = deaf of hearing parents; DWD = deaf with disabilities.

Procedures

Students were administered two receptive assessments, described below, in a one-on-one setting in the school library or a common area in the dorms. The author, who is a university professor with 8 years of pre-k-12 teaching experience and an Advanced Plus on the SLPI (Newell et al., 1983), assessed all students in Y1. Trained graduate assistants, who were students in deaf education or interpreting programs and had completed a four-course sequence of ASL classes assessed students in Years 2, 3, and 4 under the supervision of the author.

ASL Receptive Skills Test

Although the ASL-RST is limited by the small standardization sample size and the restricted age range, it is the only ASL assessment readily available to educators and researchers and the only measure with published standard scores (Enns & Herman, 2011; Enns et al., 2013). The ASL-RST was used as a measure of students’ receptive ASL skills across grammatical categories and required about 10min per participant to administer. Enns et al.’s (2013) directions specify to discontinue the assessment after five consecutive failed responses. However, because my research question addressed language skills across ages (and grammatical categories) and because of the possibility of language delays, the entire assessment was administered to all participants, including those outside the target age range (i.e., 3–13 years), in the event that students had language delays (per Enns et al., 2013). Student responses for each test item were recorded by circling the corresponding response number (i.e., 1, 2, 3, or 4) on a paper response sheet. Similar to Beal-Alvarez (2014), spatial verbs (action and location) were analyzed separately for a total of nine grammatical categories.

Receptive Test of ASL

The RT-ASL (Schick & Hoffmeister, 2001a, 2001b) was administered in Years 2 (Y2) and 3 (Y3) to investigate if students performed differently on a second receptive test. The RT-ASL combines the receptive syntax and classifier measures from previous studies (Schick, de Villiers, de Villiers, & Hoffmeister, 2007) in a 32-item assessment with one overall score. Students watch video clips presented in ASL and point to a corresponding picture in a test booklet; the researcher again circles the number corresponding to the student’s response on a score sheet. This assessment also takes about 10min to administer.

Data Analysis

To address differences in performance by various characteristics, I first calculated overall raw scores for each participant and each receptive measure. I divided overall scores for the ASL-RST and the RT-ASL and category scores for the ASL-RST into age bands (e.g., 6–7 years, 8–9 years) to present a picture of scores across ages for students without and with disabilities. I also converted the raw scores of students 6–13 years into standard scores for comparison with Enns et al.’s standard scores. I calculated Pearson’s correlation coefficients and p values for the relation between age and overall score to detect any significant correlations for each assessment.

Next, I analyzed scores by demographic characteristics. I conducted one-way between-participants ANOVAs and post hoc analyses (Tukey for equal group sizes and Scheffe for unequal group sizes) to compare scores by gender (female vs. male), parental hearing status, years attending the residential school, and year of assessment. I analyzed scores for individual DWD across measures and time. Finally, I compared performance on each receptive measure across Years 1 through 4 for a cohort of 37 students (30 DOHP and 7 DWD) who completed each assessment all 4 years.

Results

First Research Question

My first research question was: (a) How are deaf students’ receptive ASL skills (i.e., performance on the ASL-RST and the RT-ASL) affected by age, gender, parental hearing status, and additional disabilities? Table 2 shows ASL-RST raw scores by age across years for DOHP and DODP students combined and Table 3 shows results for DWD. No students, including those up to 22 years of age, scored at ceiling during any year on the ASL-RST. The highest score of 40 was obtained by one 21-year-old student in Y4. Several DOHP students, 15 years of age and older, scored 39 points (out of 42) across years, although none repeated this score. One DODP student also achieved 39 at both 12 and 13 years of age. Two DOHP students, 12;11 and 14;9, scored at ceiling on the RT-ASL in Y3. In Y4, three DOHP students (15;7 to 18;4) and one DODP (12;11) scored at ceiling. About 40% of students aged 9 years and older scored within 4 items of ceiling each year. No standard scores were available for the RT-ASL. Raw scores for all students are presented in Table 4.

Table 2.

American Sign Language Receptive Skills Test raw scores, means, and standard deviations (SDs) for age groups by test and year

Age	Y1			Y2			Y3			Y4
	N	M (DODP)	SD (DODP)	N	M (DODP)	SD (DODP)	N	M (DODP)	SD (DODP)	N	M (DODP)	SD (DODP)
5		—	—	—	—	—	—	—	—	1	23	0
6	2	17.0	8.5	0	—	—	1	24.0	0	3	12 (24.5)	0 (0.7)
7	2	20.0	11.3	1	19.0	0	1	23.0	0	1	30	0
8	1	1	17.0	4	25.5	4.2	1	30.0	0	3	20.0 (25)	9.9 (0)
9	5	25.0 (30.5)	4.4 (4.9)	1	10.0	0	3	22.7	6.8	4	30.3	2.1
10	3	20.0 (31.0)	0 (0)	6	26.3 (28.3)	5.1 (2.1)	5	26.8	4.5	3	24.7	4.2
11	2	35.0 (33.0)	0 (0)	2	26.5	3.5	3	28.0	7.2	2	27.5	6.4
12	6	31.2	2.1	4	33.3	1.0	9	31.3 (35.0)	2.9 (5.7)	9	33.0 (35.5)	2.1 (4.9)
13	8	30.8 (32.0)	3.2 (1.4)	7	34.3 (27.0)	1.0 (0)	6	29.8	4.0	3	30.0	4.0
14	5	29.5 (35.0)	2.6 (0)	9	31.8 (37.0)	2.8 (0)	1	36.0	0	4	35.3	3.3
15	6	30.2 (36.0)	2.5 (0)	5	32.5 (37.0)	4.0 (0)	6	33.3	2.9	10	33.9	4.0
16	5	361.8	2.5	6	31.7	2.1	7	32.3	2.9	8	29.5	5.4
17	11	29.3 (38.0)	4.1 (0)	7	32.6	2.9	7	33.3	1.6	9	34.8	1.8
18	9	32.9	2.5	10	30.5	5.2	7	32.9	3.0	6	32.5	5.2
19	10	31.9	3.0	5	31.8	3.6	6	32.5	2.4	3	36.0	1.0
20	7	30.9	4.2	7	34.7	4.3	10	32.1	5.8	4	31.5	4.1
21	1	29.0	0	1	32.0	0	3	33.3	4.0	4	32.8	5.6
22	1	32.0	0	0	—	—	2	33.5	3.5	—	—	—
Total	84			75			78			73

Note. (—) indicates no data. DODP = deaf of deaf parents.

Table 3.

American Sign Language Receptive Skills Test (ASL-RST) and Receptive Test of American Sign Language (RT-ASL) scores for students with disabilities across assessments and years

Age (Y1)	Gender	Disability	ASL-RST Y1	ASL-RST Y2	ASL-RST Y3	ASL-RST Y4	RT-ASL Y2	RT-ASL Y3	RT-ASL Y4
7;2	F	MID	18	17	24	27	18	19	14
7;4	M	MID	17	13	10	—	17	14	—
10;9	F	OHI	15	23	29	28	22	19	20
11;7	M	AUT	20	29	28	—	25	21	—
11;7	F	MID	21	18	16	19	18	16	14
13;8	M	MID	—	19	29	26	18	18	23
13;9	M	AUT	25	32	33	30	24	27	25
14;5	F	MOD	25	23	19	26	17	16	16
16;10	M	MOD	20	26	—	—	14	—	—
17;7	F	MID	22	20	24	23	15	—	20
18;9	F	MOD	15	21	22	23	17	—	20
19;5	F	MOD	18	19	20	—	16	—	—

Note. (—) indicates no data. AUT = autism; MID = mild intellectual disability; MOD = moderate intellectual disability; OHI = other health impairment (syndrome).

Table 4.

Receptive Test of American Sign Language raw scores by age and year

Age	Y2			Y3			Y4
	N	M (DODP)	SD (DODP)	N	M (DODP)	SD (DODP)	N	M (DODP)	SD (DODP)
5	—	—	—	—	—	—	1	14.0	0
6	—	—	—	—	—	—	3	13.0 (19.0)	0 (5.7)
7	2	20.5	6.4	—	—	—	1	16.0	0
8	3	16.0	5.6	2	24.0	2.8	2	27.0 (22.0)	0 (0)
9	—	—	—	3	21.3	5.9	4	20.0	3.4
10	7	19.3 (27.3)	4.0 (0.6)	—	—	—	3	24.0	6.2
11	2	19.0	0.4	6	24.0 (30.5)	2.9 (2.1)	2	24.0	0
12	2	26.7	2.8	2	17.5	0.7	9	26.4 (31.0)	2.4 (1.4)
13	7	27.5 (26)	2.3 (0)	4	26.3	2.6	3	25.3	2.3
14	7	26.5 (25)	3.4 (0)	4	28.5	2.9	4	28.3	2.8
15	8	26.0 (26)	2.8 (0)	6	26.2	2.2	10	27.6	3.6
16	5	26.4	1.8	4	26.0	1.8	8	28.4	2.6
17	7	27.4	3.0	4	27.5	1.7	9	25.7	2.7
18	10	25.9	3.7	6	27.3	2.6	6	27.5	3.8
19	5	19.3	0.3	9	25.7	4.5	3	24.7	1.2
20	7	26.3	0.2	3	29.5	0.7	4	23.3	3.8
21	1	29.0	0	1	29.0	1.4	3	24.7	5.0
Total	69			53			75

Note. (—) indicates no data. DODP = deaf of deaf parents.

Overall, scores on the ASL-RST and the RT-ASL tended to increase with age for students without disabilities and mean scores seemed to plateau about 5 items from ceiling at high school age. Age strongly and significantly correlated with ASL-RST scores for DOHP across all 4 years, although this correlation decreased in strength across time (Y1: N = 61, r = .634, p = .000; Y2: N = 69, r = .443, p = .000; Y3: N = 53, r = .286, p = .038; Y4: N = 71, r = .491, p = .000). For DODP, age strongly and significantly correlated with ASL-RST scores for Y1 (N = 9, r = .761, p = .009), Y2 (N = 6, r = .855, p = .015), and Y4 (N = 5, r = .895, p = .020; no data for Y3 because of only two DODP participants). Age did not significantly correlate with ASL-RST scores for DWD across any year (Y1: N = 11, r = .143, p = .676; Y2: N = 12, r = .267, p = .402; Y3: N = 11, r = .146, p = .669) except Y4 (N = 11, r = .632, p = .037). Similarly, for the RT-ASL, student age significantly and strongly correlated with scores for DOHP in Y2 (N = 69, r = .449, p = .000), Y3 (N = 53, r = .370, p = .007), and Y4 (N = 71, r = .420, p = .000) but did not correlate with scores for DODP (Y2: N = 6, r = −.005, p = .496; Y4: N = 5, r = −.709, p = .090) or DWD (Y2: N = 12, r = −.398, p = .200; Y3: N = 8, r = .346, p = .401; Y4: N = 11, r = .465, p = .465). In sum, ASL-RST scores significantly correlated with age for students with hearing and with deaf parents, whereas RT-ASL scores significantly correlated with age only for DOHP.

Standard scores

Next I compared students’ overall performance on the ASL-RST to their same-age native or near-native signing peers from Enns et al.’s (2103) standardization sample (i.e., 6–13 years of age). DOHP scores fell within the average range for their age (i.e., ±1 SD below or above the mean; Enns et al., 2013) across all 4 years with the exception of one student in Y4. DWD students within the standardization age range scored in the low average to average range with the exceptions of 1–2 students per year.

Gender

Next, I investigated gender effects for each assessment. I conducted a one-way between-participants ANOVA to investigate gender effects for all DOHP girls and boys without disabilities across 4 years of ASL-RST scores and 3 years of RT-ASL scores (only one DODP was female; therefore, I did not investigate gender effects relative to the DODP group). There was no significant effect for gender in any year for the ASL-RST or the RT-ASL (see Table 5).

Table 5.

DOHP mean raw score and standard deviation (SD) by gender for the 42-item ASL-RST and the 32-item RT-ASL

	ASL-RST								RT-ASL
	Y1		Y2		Y3		Y4		Y2		Y3		Y4
	M	F	M	F	M	F	M	F	M	F	M	F	M	F
N	33	28	39	30	30	23	39	32	39	30	30	23	39	32
M	30.2	28.5	31.9	30.0	32.5	31.0	31.8	30.9	25.5	24.4	26.3	25.1	26.1	24.6
SD	4.9	6.0	5.1	4.9	3.5	4.1	5.2	6.0	4.4	4.3	3.3	4.3	3.8	4.8
p	.227		.129		.176		.493		.303		.176		.192

Note. M = male; F = female; ASL-RST = American Sign Language Receptive Skills Test; DOHP = deaf of hearing parents; RT-ASL = Receptive Test of American Sign Language.

A total of 136 different students were assessed across 4 years (including 16 DWD and 12 DODP). Most students had attended the residential school for 1 (24%) or 2 years (18%) at the time of testing. About 10% of students attended the school for 3, 4, and 5 years. An average of 4% of students attended for each of 6 through 12 years. I analyzed annual ASL-RST and RT-ASL scores for DOHP by years of attendance, which ranged from 1 to 12, as a proxy for minimum number of years of exposure to ASL. With students of all ages included, there was no significant difference between years of attendance and ASL-RST mean scores across any year (Y1: [F(8, 55) = 2.04, p = .59]; Y2: [F(8, 54) = 1.13, p = .357]; Y3: [F(10, 62) = 1.19, p = .315]; Y4 [F(9, 64) = 1.73, p = .100]). Similarly, there was no difference between years of attendance and RT-ASL mean scores for any year (Y2: [F(8, 54) = .916, p = .51]; Y3: [F(10, 61) = 1.08, p = .39]; Y4: [F(9, 62) = .730, p = .68]). Because the numbers of students within each group varied from 3 to 18, I combined individual years of attendance into four groups: 1–2 years, 3–4 years, 5–6 years, and 7-plus years of attendance at the school, which created four groups of 9–33 students each. Again, there was no significant difference between ASL-RST mean scores and years of attendance group for Y1 (F(3, 60) = 2.73, p = .052), Y2 (F(3, 59) = 2.18, p = .10), Y3 (F(3, 69) = 1.37, p = .26), or Y4 (F(3, 70) = 1.35, p = .27) or for the RT-ASL for Y2 (F(3, 59) = .930, p = .43), Y3 (F(3, 68) = 1.32, p = .27), or Y4 (F(3, 68) = 1.72, p = .17).

Finally, given the large population of middle and high school students, I analyzed performance related to years of attendance for the youngest group of students, who were 6;4 to 11;7 and attended the school for 1–7 years at each testing year. There was a significant difference between ASL-RST scores and years of attendance for this group (F(6, 54) = 3.07, p = .012), but Scheffe post hoc testing revealed no statistically significant difference between any 2 years of attendance. There was a moderate and significant correlation among years of attendance and ASL-RST scores for this group (r = .352, p = .005). In contrast, there was no significant difference between years of attendance and RT-ASL scores (F(6, 33) = 1.59, p = .182) and the correlation between these two factors was not significant for this young group (r = .022, p = .362).

Second Research Question

My second research question was: How do students’ receptive ASL skills change across time? Longitudinally across four academic years, 30 DOHP, 2 DODP, and 6 DWD completed the ASL-RST. From Y1 to Y2, 19 DOHP (63%) increased their scores (by 1–11 items), 2 (7%) scored the same, and 9 (30%) decreased in score (by 1–7 items). From Y2 to Y3, 18 (60%) increased their score (by 1–9 items), 4 (13%) scored the same, and 8 (27%) decreased in score (by 1–11 items). Finally, from Y3 to Y4, 24 (80%) increased their score (by 1–9 items), 1 scored the same (3%), and 5 decreased (17%; most by 1 item; two students by 6 and 17 items). Longitudinally across 4 years, there was a significant difference in the mean ASL-RST performance of the DOHP cohort (F(3, 116) = 6.68, p = .000). Mean scores in Y3 (M = 30.1, SD = 4.1; p = .027) and Y4 (M = 33.0, SD = 5.2; p = .000) were significantly higher than Y1 scores (M = 27.1, SD = 6.4). According to these results, students needed at least two school years of instruction as a group to significantly increase their mean score by about 3 items.

For the RT-ASL, 14 students (48%) increased their performance from Y2 to Y3 (by 1–8 items), 6 (21%) scored the same, and 9 (31%) decreased (by 1–5 items; one student did not take the RT-ASL in Y3 for a total of 29 students). From Y3 to Y4, 19 (66%) increased in score (by 1–10 items), 3 (10%) scored the same, and 7 (24%) decreased (by 1–6 items). There was no significant difference in RT-ASL mean scores by year for the 30 DOHP students (F(2, 86) = 1.48, p = .233).

The two DODP students increased their scores by 3 and 9 items across the 4-year period for the ASL-RST and by 4 and 5 items for the RT-ASL. Results for DWD are shown in Table 3. Half of the DWD increased and half decreased across each year of the ASL-RST and the RT-ASL, usually by 1–4 items. There was no significant difference in their mean score by year for the ASL-RST (F(3, 20) = 1.77, p = .186) or the RT-ASL (F(2, 13) = .166, p = .848). Figure 1 presents ASL-RST performance by group and age cohort across 4 years.

Figure 1.

American Sign Language Receptive Skills Test group raw score mean for 30 students across 4 years by age cohort (out of 42 items). Note. DWD = deaf with disabilities; DODP = deaf of deaf parents.

Grammatical categories

Next I looked at the grammatical category performance across 2-year age bands within the cohort of 30 DOHP and 6 DWD students (see Tables 6 and 7). In general, mean scores increased with age. Older students performed higher than younger students for most categories but with flatter slopes and younger students’ achievement gap decreased as they approached the performance of older students. DWD scored lower than their peers with a flat slope across time (see Figure 1). Most age groups increased across most categories all 4 years. However, in some areas, scores seemed to plateau or decrease from beyond Y2 for some age groups, such as number-distribution, location, and role shift (although role shift is limited by only 3 items compared to 7 and 8 items for the other two categories). DWD showed increases in some areas (e.g., action) and plateaus in others (i.e., number-distribution, noun-verb, SASS, conditionals). Location appeared to be a most difficult category, with the highest group mean score 2 items below ceiling. No age group scored at ceiling for any category during any year.

Table 6.

Raw scores for age group cohorts across 4 years and five categories of the American Sign Language Receptive Skills Test

Y1 Age	N	Number-distribution				Negation				Noun-verb				Location
		Y1	Y2	Y3	Y4	Y1	Y2	Y3	Y4	Y1	Y2	Y3	Y4	Y1	Y2	Y3	Y4
6–7	3	2.3	5.0	4.3	3.3	2.0	5.3	6.0	4.7	2.3	3.0	3.0	3.0	3.0	3.3	4.3	4.7
8–9	4	3.3	4.0	4.3	5.0	4.8	5.0	6.0	6.5	2.3	2.5	3.3	3.8	4.0	4.8	4.8	5.5
10–11	3	4.0	4.3	4.3	5.0	4.7	7.0	7.3	7.3	3.0	3.0	3.3	3.3	5.7	4.7	4.3	6.0
12–13	7	4.6	5.7	5.0	6.4	7.1	7.7	7.9	8.3	3.6	3.1	3.4	3.7	5.0	5.6	4.9	5.7
14–15	5	4.2	5.2	5.8	5.8	6.2	7.6	7.4	7.8	2.6	3.2	3.4	3.6	5.4	5.2	5.4	6.4
16–18	8	4.5	5.0	4.0	6.0	6.0	7.0	5.0	7.4	2.6	2.9	2.0	3.1	5.0	6.1	4.7	4.7
DWD	7	3.0	3.0	4.0	3.0	3.1	4.5	5.0	5.1	2.1	2.3	2.6	2.6	4.3	3.8	5.4	4.9

Note. DWD = deaf with disabilities.

Table 7.

Raw scores for age group cohorts across 4 years and four categories of the American Sign Language Receptive Skills Test

Y1 Age	N	Action				SASS				Handle				Role shift				Conditionals
		Y1	Y2	Y3	Y4	Y1	Y2	Y3	Y4	Y1	Y2	Y3	Y4	Y1	Y2	Y3	Y4	Y1	Y2	Y3	Y4
6–7	3	3.3	5.2	6.7	6.3	1.0	2.7	2.7	2.7	1.3	1.7	2.7	2.7	1.3	1.3	1.3	2.3	0.7	1.0	1.0	1.0
8–9	4	5.3	5.0	6.0	6.0	2.0	2.5	3.5	3.3	2.3	1.8	2.5	2.0	1.3	1.5	2.3	2.0	1.0	1.3	1.8	1.8
10–11	3	4.7	5.0	5.7	6.3	1.7	3.7	3.0	3.0	1.3	2.7	2.0	2.7	1.3	2.0	1.3	1.3	1.3	2.0	1.3	1.3
12–13	7	6.4	6.7	6.4	7.4	3.0	3.0	3.3	3.4	2.3	2.3	2.4	2.6	2.3	2.1	1.7	2.4	1.7	1.9	1.9	1.9
14–15	5	6.6	5.8	7.0	7.2	2.2	3.2	3.2	3.6	2.2	1.8	2.4	2.8	1.6	1.2	1.6	1.6	2.0	2.0	2.0	2.0
16–18	8	6.1	6.4	5.3	7.0	2.5	3.3	2.7	3.3	2.4	2.6	1.3	2.4	1.6	2.1	2.0	2.3	2.0	1.7	1.0	2.0
DWD	7	4.1	4.9	5.7	5.0	2.0	2.5	2.1	2.1	1.7	2.0	2.3	2.3	1.7	1.1	1.1	2.0	1.3	1.1	1.3	1.4

Note. DWD = deaf with disabilities; SASS = size-and-shape-specifier classifiers.

Similar to Beal-Alvarez (2014), I conducted an error analysis across each of 4 years for the cohort of 30 DOHP who completed the ASL-RST (see Table 8). Items are included if one third or more of students responded incorrectly for an item in Y1, which occurred for half of the test items, with most of these items at the end of the test. Missed items were spread across all nine grammatical categories, with 5 items including two categories in the stimulus. The first six listed items show a decrease in errors across time, as do items 30, 34, 39, and 41. Student errors fluctuated across other items but more than one third of students missed the last 11 test items across all 4 years.

Table 8.

Percentage of student errors by item number and year on the American Sign Language Receptive Skills Test for a cohort of 30 DOHP students who completed the test across each of 4 years

Item	Y1	Y2	Y3	Y4	Category
2	30	10	3	0	SASS
6	50	30	23	23	Negation
8	37	23	17	10	Negation
9	43	27	20	13	Role shift
10	37	20	10	10	SASS
18	43	17	17	7	Number-distribution
23	77	60	70	50	Number-distribution
26	43	23	33	23	Spatial verb (action)
28	33	57	33	17	Spatial verb (action) and handling CL
29	50	47	47	33	Spatial verb (location)
30	40	33	30	20	Spatial verb (action) and handling CL
31	70	57	63	40	Negation
32	70	73	57	47	Noun-verb
34	47	37	33	33	Negation and conditional
35	70	57	60	40	Number-distribution
36	40	53	57	43	Role shift
37	67	73	63	50	Spatial verb (location)
38	73	50	63	57	Negation and SASS
39	80	70	70	53	Spatial verb (location)
40	53	57	57	50	Role shift
41	60	53	50	33	Number-distribution and spatial verb (action)
42	77	80	80	63	Spatial verb (location)

Note. DOHP = deaf of hearing parents; SASS = size-and-shape-specifier classifiers.

Finally, I investigated student performance across time between receptive measures. I converted overall ASL-RST and RT-ASL raw scores to percentages because the assessments had a different number of total items. More DOHP scored consistently higher across the 3 years on the RT-ASL than on the ASL-RST. At Y2, 50 (74%) scored higher on the RT-ASL than the ASL-RST (15; 22%), and 3 (4%) scored the same. Again in Y3, 50 (68%) scored higher on the RT-ASL, 23 (31%) scored higher on the ASL-RST, and 1 scored the same. In Y4, 47 (67%) scored higher on the RT-ASL and 23 (33%) scored higher on the ASL-RST. However, 12–16 students per year scored within 3% on each measure annually. Nearly all DODP scored higher on the RT-ASL across all years, whereas the DWD were evenly divided on performance across years. Half scored higher on the RT-ASL and the other on the ASL-RST. Finally, ASL-RST and RT-ASL scores strongly and significantly correlated across Y2 (N = 63, r = .793, p = .000) and Y3 (N = 73, r = .774, p = .000) but not Y4 (N = 72, r = −.013, p = .456). Below I discuss these findings.

Discussion

The present study provides longitudinal empirical receptive ASL data across a student body at a residential school, including those who are DWD. These data were previously unavailable in the literature. In general, receptive ASL scores increased with age for DOHP, similar to Beal-Alvarez (2014), but not for the smaller samples of DODP or DWD. Strength of this correlation decreased slightly across time, suggesting possible repeated testing effects. Students within the age range for standard scores performed similarly to their peers (Enns et al., 2013). They also scored within average standard score ranges across time, which extends previous findings (Beal-Alvarez, 2014) to longitudinal results. DWD scored in the average to low average standard score range and consistently lower than their peers with an increasing gap across time, which is in contrast to the DWD students in Mann et al.’s study (2013), although the present sample of DWD contained primarily those with intellectual disabilities. No differences were found for gender on either receptive measure, similar to previous findings (Haug, 2011; Maller et al., 1999) and different from others (Hermans et al., 2010; Herman & Roy, 2006).

Older students showed little movement across time. It appears the ASL-RST maxed out around 11–12 years of age, as older students made little improvement across time, yielding largely uninterpretable results. Significant changes in performance, at a mean increase of 3 items on the ASL-RST, required a period of 2 years. Similarly, with the exception of a few students, overall students did not reach ceiling on the RT-ASL, even at the oldest ages. These data align with previous results from a school for the deaf with a bilingual program that showed students in the process of acquiring a first language need additional time to show developmental progress in both language and academic abilities (Lange et al., 2013), even while immersed in a bilingual educational environment. In the present study, years of attendance at the residential school did not significantly relate to performance on either assessment overall, although the correlation between years of school attendance and ASL-RST scores was significant for younger students.

One might speculate that deaf children in a signing environment would consistently increase their language skills with time. The present results raise the question of why this prediction in general was not bore out by the current longitudinal receptive language data of students as measured by these two assessments. Three possibilities come to mind. First is the factor of student age at first exposure to consistent, fluent models of ASL and practice opportunities. Some deaf students are exposed to sign language at later ages (Andrews & Covell, 2006/2007; Mayberry & Eichen, 1991; Musselman & Tane Akamatsu, 1999) and samples of late learners of ASL have shown limitations in their sign language proficiency (Mayberry & Eichen, 1991). It is possible that many of the DOHP in the current study were late learners of ASL, considering that the majority of students were high school age and were first- or second-year students at the research site during longitudinal assessments. Additionally, fluctuation in attendance at the school demonstrates a pattern of inconsistent exposure to sign language-rich environments.

Another possibility related to the current results is the consistency and quality of ASL models for instruction and practice within the educational setting. In general, the majority of deaf education teachers and educational interpreters are females with typical hearing who learned ASL during their university preparation programs (Corbett & Jensema, 1981; Krause, Kegl, & Schick, 2008; Stauffer, 2011; Storey & Jamieson, 2004; van Dijk, Boers, Christoffels, & Hermans, 2011; Woodward, Allen, & Schildroth, 1988; Yarger, 2001). It is possible, then, that most deaf students are exposed to language models who learned ASL as a second language in adulthood. Although no data are available for ASL skill levels of teachers of the deaf (Beal-Alvarez & Scheetz, 2015), previous studies showed that educational interpreters are in need of continuous professional development to hone their ASL skills beyond their preparation programs (Dean & Pollard, 2001; Schick, Williams, & Bolster, 1999; Storey & Jamieson, 2004). Recently, Beal-Alvarez and Scheetz (2015) reported interpreter and teacher candidates at the university level were accurate in their own ratings of their ASL skills compared to their professor’s ratings and candidates accurately rated the skills of child signers but needed more exposure to less fluent signers and instruction on actual ASL features (e.g., pronominalization). Some schools for the deaf have established required ASL proficiency levels for teachers, such as a 3+ on the ASLPI for “acceptable” (see http://www.gallaudet.edu/asldes/aslpi/aslpi_proficiency_levels.html for level description; S. Shapiro, personal communication, December 11, 2014), or an “advanced” rating on the SLPI (L. Jackson, personal communication, December 23, 2014). However, public schools frequently do not establish required proficiency levels (Beal-Alvarez & Scheetz, 2015).

Currently, schools for the deaf tend to have higher percentages of deaf teachers than local schools (Allen & Karchmer, 1990; Andrews & Franklin, 1996/1997), which presents an ideal educational situation for those students who use sign language. However, prevalence of deaf teachers remains low, with estimates at 20–30% (Allen & Karchmer, 1990). At the present school, eight teachers were deaf. Six deaf teachers taught at the high school level, one at the middle school level, and one was the ASL teacher who served all grades. Although high schoolers frequently interacted with deaf teachers, younger students infrequently came into contact with deaf native or native-like signing adults during the early ages of language acquisition. This highlights the need for recruitment of more deaf teachers in the field of deaf education.

Students also need practice opportunities in ASL. To date, researchers have demonstrated that deaf children with stronger ASL skills have stronger English skills (Bochner & Albertini, 1988; Chamberlain & Mayberry, 2000; Hoffmeister, de Villiers, Engen, & Topol, 1998; McQuarrie & Abbott, 2008; Padden & Ramsey, 1998; Prinz & Strong, 1998) and advocated ASL-English bilingual programs (Bailes, 2001; Berke, 2013; Crume, 2013; Lange et al., 2013; Mashie, 1995; Padden & Ramsey, 1998; Strong, 1995); however, data are unavailable to guide actual implementation of bilingual instruction at the pre-k-12 level. When do teachers address which elements in ASL instruction? How do they document a child’s current ASL proficiency level and use progress monitoring to guide their instruction? Presently, ASL instructional standards at the k-12 level, similar to Common Core State Standards for English/Language Arts by grade level, are unavailable, although Ashton and colleagues (2014) presented goals and related standards for how k-16 students should be able to use ASL and Swaney (2015) provided a review on curricula and materials used by ASL instructors. Educators need specific standards to drive instruction and students, both typically hearing and deaf, should have access to sign language courses across primary and secondary levels, including deaf students who use ASL as a first language in tandem with English Language Arts and those who learn it for foreign language credit at the secondary level, including possible future teachers and interpreters (Beal-Alvarez & Scheetz, 2015), to remediate current models of interpreter and teacher preparation programs in the United States that require no ASL skills upon program entry (Humphries & Allen, 2008; Napier, 2004).

A third possibility related to students’ ASL scores is the assessment administration process. The premise for each receptive assessment is that when a child sees the signed stimulus, s/he immediately identifies the correct picture of four choices. For the ASL-RST, each response screen is displayed immediately after each stimulus for approximately 3s. For the RT-ASL, students watch the signed stimulus on a computer screen and then select either a distinct picture from three or four options or a specific element within one picture. For example, the signer describes the position of four animals and points to one of them. All animals are in the same picture, as opposed to four different pictures. Pictures are displayed in a book directly below the computer screen. It is possible that students had more time to look at the RT-ASL responses, as timing was not embedded into the response presentation. In the present study, items were not repeated unless requested by a student or if the researcher noticed a visual distraction during a test item. Overall, repetitions of test items were minimal. It is possible students missed signed stimuli for either measure or that they missed the response choices for the ASL-RST. These differences in administration may have affected the number of correct responses, as most students performed consistently higher on the RT-ASL. On the other hand, guessing on responses (each response had a 25% chance of guessing correctly) and repeated testing may have contributed to student performance. Future video-recording of student behavior during the assessments might speak to the frequency of student distraction, repetitions, and their possible effects on test performance.

A final possibility affecting student performance across time is the specific test items. The ASL-RST and the RT-ASL are young assessments with limited published data on student performance. ASL-RST items increase in difficulty by item number, which is confirmed in the present and previous samples (Beal-Alvarez, 2014; Enns et al., 2013). The expectation is that children learn these structures by the age of 13 years, although this has not been confirmed in present or previous test administrations (Beal-Alvarez, 2014). Future investigations might analyze how proficient deaf adult signers respond to these assessments, similar to Haug’s (2011) work, for comparison with the present error analysis across items. This information might highlight test stimuli in need of revision.

Conclusion

The current data address the call for longitudinal profiles of students’ receptive ASL skills across 4 years for two measures (Allen & Enns, 2013; Beal-Alvarez, 2014; Hermans et al., 2010; Johnson, 2004; Mann et al., 2013) and provide educators and researchers with a starting point for understanding how the diverse deaf student population performs longitudinally on two receptive ASL measures. Educators can compare their students’ scores to current data to address how a similar student (e.g., DOHP, DODP, or DWD) might be expected to develop over time. Although the specifics relative to participants’ language experiences and age of acquisition are unknown, in general, children’s skills increased very little across 4 years, despite explicit ASL instruction as an elective course, the provision of instruction via sign language, and access to communicate partners outside of instructional time. A residential school that utilizes a bilingual ASL/English approach appears to be the most accessible educational environment for a child who relies on ASL for communication and instruction; yet as a group these students advanced relatively little over time on these measures and did not achieve ceiling scores, despite their ages beyond 13 years. Perhaps schools and districts who serve deaf children should add ASL/sign language specialists to their staff to annually assess and evaluate students’ sign language skills and to make data-driven instructional decisions in line with legislative (e.g., IDEIA, 2004) and teacher training initiatives (e.g., edTPA; SCALE, 2013). Additionally, instruction specific to a signed language arts curriculum, based on sign language development, should be employed at bilingual schools for explicit language instruction. These empirical data support the need for more effective early intervention and language instruction for deaf children, for both those who use sign language and those who use spoken language (Mayer & Leigh, 2010).

Conflicts of Interest

No conflicts of interest were reported.