German Sign Language Vocabulary Test (DGST)
The German Sign Language Vocabulary Test (DGST; Mann, Temmen & Klinner, 2024) is an adaptation of the British Sign Language Vocabulary Test (BSL-VT; Mann, 2009; Mann & Marshall, 2012). It is used to assess the vocabulary knowledge of deaf children by “measuring the strength of the associations between the form and meaning of items in the core lexicon” (Mann & Marshall, 2012, p. 1031). The DGST was developed in close collaboration with a panel of deaf and hearing DGS experts.
Development of the instrument
Like the web-based BSL Vocabulary Test, the DGST consists of four tasks that measure vocabulary knowledge: (1) meaning recognition, (2) form recognition, (3) meaning recall, and (4) form recall. The same items are used across all tasks. Like the BSL Vocabulary Test, the DGST is based on Laufer and colleagues' model of Strength of Vocabulary Knowledge (Laufer, Elder, Hill & Congdon, 2004; Laufer & Goldstein, 2004). This model suggests that spoken vocabulary develops incrementally and that there is a hierarchy in the degree of strength of vocabulary knowledge. Each of the four DGST tasks consists of 120 vocabulary items. The test includes two receptive and two production tasks. The two receptive tasks use a multiple-choice format and can be self-administered if the test taker is old enough to operate a mouse or touchscreen. The two production tasks require an administrator to score each response, as well as documenting the response in a text box on the computer screen using German glosses.
The task formats are as follows: Meaning Recognition Task: test takers see a pre-recorded DGST sign followed by four pictures. They have to select the picture that corresponds to the meaning of the signed prompt (Video 1). In the Form Recognition Task, test takers see a picture, followed by four pre-recorded DGST signs and have to select the sign that matches the meaning of the picture prompt (Video 2). In the Form Recall Task, test takers see a picture and have to produce the corresponding DGST sign. Finally, in the Meaning Recall Task, children see a pre-recorded DGST sign and have to generate three DGST signs with associated meanings.
A pre-pilot and pilot of the completed set of test terms were carried out before finalizing the prototype of the adapted DGST vocabulary test. In the pre-pilot, a native signer completed all vocabulary tasks and provided feedback on any items they considered problematic and/or suggestions of format-related changes. This was done as an additional measure to ensure test validity. Based on the feedback, any final revisions were made and uploaded to the test website. In addition, a PowerPoint version of the test was developed to allow for more flexible administration in areas where internet access is problematic.
Forty-one children (19 boys and 22 girls) between the ages of 4-15 years (M=10;5, SD=3.3) participated in the pilot, including three CODAs.
Psychometric information (from Mann, Temmen, & Klinner, under review)
Reliability for internal consistency (Cronbach’s alpha)
- meaning recall (.99)
- form recall (.97)
- form recognition (.98)
- meaning recognition (.98)
Item analysis
-3 items (FISH, CAR, CAP) passed by all participants on three of the four levels of vocabulary knowledge (meaning recognition, form recognition, and form recall)
- none of these items were passed by all participants on the meaning recall task.
- no item were failed by all participants across all four tasks
Inter-rater reliability (Cohen's kappa)
- Agreements between raters’ judgments for scoring the form recall task (k=.92)
- Agreements between raters’ judgments for scoring the meaning recall task (k=.89)
- Items from the two receptive tasks were automatically scored by the computer upon selection of the response.
Content validity
Content validity was established in the following ways.
- Close collaboration with a panel of deaf and hearing experts on the development of the target and distractor items.
- External validity was ensured by an even distribution of items on the MoA (Mode of Acquisition) scale, as assessed by deaf teachers. Mode of Acquisition (MoA, Wauters et al., 2008) ratings of target items reflect the range/spread of item types from concrete to abstract signs: spread of ratings over most of the range (1-5) except ‘5’ (signs that are acquired exclusively through language)
- To validate the image material, the drawings for all items were checked for clarity by a child in the tested age group. This led to changes of 22 drawings by the deaf artist.
- To validate the video material, all videos were reviewed by two native speaker of German Sign Language. This led to the replacement of 14 videos.
Construct validity
- Correlation analysis between participants’ performance on the different tasks and age:
- age & meaning recognition = 0.91, p<.001
- age & form recognition = 0.90, p<.001
- age & form recall = 0.86, p<.001
- age & meaning recall = 0.77, p<.001
Analysis of differences between participants’ performances across tasks:
Main effect of task, F (2.301, 92) = 337.781, p<.001, ηρ. = .894
Form recall > meaning recall (p<.001)
Meaning recognition > form recognition (p<.001)
Meaning recognition > meaning recall (p<.001)
Form recognition > form recall (p<.001)
Availability
The DGST is currently available to researchers upon request.