Vocabulary instruction for students with learning disabilities: a review of the research.

Abstract. This article reviews research on vocabulary interventions involving students with learning disabilities. The purpose of the review was to summarize the findings of vocabulary intervention research and to present implications for vocabulary instruction. The studies were analyzed based on the following intervention research variables: word selection procedure, materials, instructional design and procedures, duration of the intervention, mastery criterion, measures of vocabulary learning (word knowledge level and reading comprehension), maintenance, and generalization (reading comprehension). Based on the selection criteria, six articles spanning the period from 1978 to the present were identified. Vocabulary interventions were categorized into four areas, computer-assisted instruction (CAI), fluency-building vocabulary practice activities, mnemonic strategy instruction, and concept enhancement instruction. Overall, positive results were found for the interventions on measures of immediate recall, maintenance, and generalization.

**********

With increased reading requirements at the secondary level, students encounter extensive content-area vocabulary that includes specialized meanings that students must understand to comprehend subject-area text. Thus, knowing the meanings, relationships, and contextual interpretations of new vocabulary words enhances comprehension of content-area text (Baumann & Kameenui, 1991). Activities such as wide reading and the use of independent word-learning strategies facilitate the learning of word meanings and an understanding of concepts, which are pervasive in content-area reading. However, secondary students with learning disabilities (LD) in reading typically do not engage sufficiently in either activity, and consequently possess widely disparate vocabulary knowledge compared to their typically achieving peers (Baker, Simmons, & Kameenui, 1995). Yet, vocabulary knowledge "is not an all-or-nothing proposition; it is not the case that one either knows or does not know a word" (Beck & McKeown, 1991, p. 791). Rather, there are different levels of processing word knowledge, including association, comprehension, and generation, depending on the purpose of vocabulary instruction (see Stahl, 1986). The challenge is to identify methods that effectively teach students with LD how to process and comprehend unknown word meanings.

Researchers have sought to identify effective methods for teaching new vocabulary that focus on the word-acquisition level. Traditional independent word-learning strategies have consisted of dictionary usage and context clues.

Dictionary usage involves multiple skills such as using guidewords, decoding, discerning the correct definition, and so forth. Because of the complexities of dictionary use (e.g., multiple definitions, unknown words), struggling students may not benefit from using this strategy exclusively (Nagy & Stahl, 2000). McKeown (1993) found that the nature of definitions (e.g., rewritten or elaborated versus taken directly from the dictionary) affects students' ability to comprehend and apply word meanings. For example, drawing from associative-memory facilitation learning, the keyword strategy, which provides a mnemonic for retention, has been shown to help struggling students learn and remember key content-area vocabulary definitions (Mastropieri, Scruggs, Levin, Gaffney, & McLoone, 1985).

Context clues, another independent word-learning strategy, presents its own challenges. For example, Beck and McKeown (1991) noted that using context to derive word meanings is a challenging process that involves integrating different types of information (e.g., definitions, examples, and synonyms) from text to figure out unknown vocabulary. Moreover, using context clues to comprehend new words may be helpful only across time after multiple encounters with words (Nagy & Stahl, 2000). Overall, results are equivocal for teaching struggling students how to use context to derive meanings of new words (Carnine, Kameenui, & Coyle, 1984; Patberg & Stibbe, 1985).

At best, definitional and contextual strategies should be combined, and explicit instruction (i.e., modeling, multiple practice opportunities, and guided practice) is recommended for teaching not only the use of these strategies but also the words deemed critically important to enhance that comprehension. Additionally, teacher-led instruction has been recommended for teaching concepts associated with content-area material.

Instruction in the conceptual understanding of word meanings, their semantic relationships to other concepts, and linkages to prior knowledge has been identified as important for teaching vocabulary at a deeper word processing level for text comprehension (Pressley & Woloshyn, 1995). For example, research findings have shown that a relationship technique, such as semantic mapping (Pittelman, Levin, & Johnson, 1985), can help students to examine the relationships and associations between concepts and to develop deeper understandings of word meanings compared to students who studied words using a contextual strategy (Margosein, Pascarella, & Pflaum, 1982). Overall, research findings suggest that there is no one best way to teach vocabulary (Baker et al., 1995); rather, using a variety of techniques that include multiple exposures to unknown word meanings produces the best results (Beck & McKeown, 1991).

The goal of vocabulary instruction is to facilitate students' ability to interact with language situations, most notably comprehending text (Beck & McKeown, 1991). Vocabulary knowledge is fundamental to comprehending text, as evidenced in studies that have found a relationship between vocabulary knowledge and comprehension (Bos & Anders, 1990). Findings that support the connection between vocabulary instruction and reading comprehension point to multiple practice opportunities, indepth word-meaning knowledge, and active processing of information as important components of instruction (Stahl & Fairbanks, 1986). Nagy (1988) contended that deep word-knowledge learning is accomplished by helping students link or integrate meanings of new words to prior learning, providing multiple repetitions to aid students' retention of new meanings, and engaging students in meaningful ways that require applications of word meanings across situations.

Vocabulary instruction that produces indepth word knowledge and increases reading comprehension is necessary for secondary students with LD (Simmons & Kameenui, 1990) to manage the reading demands of secondary instruction. Converging evidence suggests that while vocabulary acquisition is positively affected by the wealth of reading opportunities, it is negatively affected by memory and language deficiencies and by limited independent word learning strategies (Baker et al., 1995), which are typical difficulties manifested by students with LD. Thus, instructional techniques must focus on ways to enhance retention of new vocabulary at the word-meaning level and to transfer this knowledge so as to assist with the comprehension of text. Students must learn strategies for independently developing a deeper understanding of the conceptual meanings of words that often constitute vocabulary found in content-area text. Recognizing that it takes time for students with LD to learn strategies, explicit instruction and opportunities to practice are important components of vocabulary instruction.

Research studies on vocabulary instruction for students with LD have been conducted, and Baker et al. (1995) provided a curriculum design framework to support vocabulary instruction for diverse learners. However, a review of the research findings on vocabulary instruction is lacking for students with LD.

The purpose of this article was to summarize the findings of vocabulary intervention research designed for students with LD and to present implications for vocabulary instruction.

METHOD

Literature Search Procedures and Selection Criteria

Studies reviewed in this article were identified from special education journals from 1978 to the present. Specifically, a computer-assisted search was conducted on the Psychological Abstracts (PSYCLIT) and Educational Resources Information Clearinghouse (ERIC) databases using the following descriptors, "learning disabilities," "vocabulary interventions," and "vocabulary instruction." Subsequently, several criteria were used to select the articles that were included in the final review.

First, participants had to be identified as middle or high school-aged students with an identified learning disability. Articles focusing only on participants who were identified as having a mild disability and/or as being at-risk or remedial students were excluded. Finally, articles that involved students with LD and students with other types of disabilities were included only if the results were disaggregated for students with LD. Second, articles had to report on intervention studies with the purpose of teaching vocabulary to students with LD. Thus, the dependent measure had to focus on the effects of the vocabulary intervention. Studies that included separate measures for vocabulary and reading comprehension were included. Third, only published studies were selected.

Independently, the authors screened each article to ensure that it met the criteria for inclusion. Agreement on articles that met the criteria was established and articles were either included or omitted.

A hand search of journals was not conducted, and dissertations and book chapters were not included; thus, conclusions based on this review should be viewed as tentative. The journals included The Journal of Special Education, Learning Disability Quarterly, Journal of Learning Disabilities, Learning Disabilities Focus, Learning Disabilities Research, Learning Disabilities Research & Practice, Exceptional Children, Remedial and Special Education, and British Journal of Educational Technology.

INTERVENTION STUDIES

The research studies to be discussed are organized by intervention categories. The authors independently reviewed each article and determined the intervention category before meeting to confirm their categories. The studies were examined according to intervention research variables, including word selection procedure, materials, instructional design and procedures, duration of the intervention, mastery criterion, measures of vocabulary learning (word-knowledge level and reading comprehension), maintenance, and generalization to reading comprehension. The categories of vocabulary interventions include computer-assisted instruction (CAI), fluency-building vocabulary practice activities, mnemonic strategy instruction, and concept enhancement instruction.

Computer-Assisted Instruction (CAI)

Johnson, Gersten, and Carnine (1987) used CAI as a means of delivering vocabulary instruction independent of teacher-directed instruction (see Table 1). The authors studied the effects of size of instructional set (i.e., number of vocabulary words presented) and practice routines when delivered through a CAI format. Twenty-five students in grades 9-12 were matched according to pretest scores on vocabulary achievement and randomly assigned to one of two treatments, the small teaching set (STS, 7 words) or large teaching set (LTS, 25 words) group, in a resource setting where students worked individually on computers.

In the STS, a bank of unknown vocabulary words was identified and individualized for each student. The practice exercises consisted of three types of multiple-choice items: the new word appeared alone with the correct definition as one of five choices, the word appeared in a sentence with the correct definition as one of five choices, or a synonym (or short definition for the word) appeared in a sentence with the word as one of five choices. The criterion for mastery was identifying the word's meaning correctly four times in a row across two lessons. A cumulative review test on 10 mastered words was also part of the program. Missed words were put back in the practice set, and the student had to again meet mastery criterion on the word.

In the LTS, two sets of 25 words were presented. These words were not individualized; therefore, some of the students knew the meaning of some of the words at the onset. The program presented a menu with a choice of four formats: word display, multiple-choice quiz, sentence completion, or an arcade-type game. Two word displays and multiple-choice quizzes were similar to the STS program. Sentence completion involved spelling the new words; the arcade activity involved matching exercises in a game format. Students were told they could play an arcade game when they reached mastery of 84% correct and could move on to a new activity after scoring 84% correct for two consecutive days. After meeting the criterion of two days at 84% correct, the students began the second list of 25 words, following the same sequence of activities. Completion of all activities on both word sets constituted mastery for the study.

The same 50 words were taught in both conditions over the course of 11 sessions lasting 20 minutes each. Results indicated that, given 11 instructional sessions, students in the STS outperformed students in the LTS in terms of time required to reach the mastery criterion, and maintenance results showed only a slight decrease in percentage correct performance for both groups. However, on transfer measures, which differed in format (i.e., open-ended oral test and passage comprehension) from the multiple-choice criterion-referenced post- and maintenance tests, students scored poorly on both the STS and the LTS.

Fluency-Building Vocabulary Practice Activities

One paper was identified that can be categorized as a fluency-building practice activity (see Table 2). Stump, Lovitt, Fister, Kemp, Moore, and Schroeder (1992) conducted two studies (Study 2 replicated Study 1) to examine the effects of a vocabulary fluency-building activity on students' learning and retention of vocabulary selected by teachers from their content-area material. Both studies used the same single-subject design, which consisted of a baseline phase where teachers introduced and explained the unit vocabulary and related concepts to the entire class. Students then studied the vocabulary words independently for 5-10 minutes. Next, students were given 3 minutes to complete as many items as possible on a quiz. Students exchanged and corrected quizzes, and the data (correct and incorrect) were recorded on student data sheets. Items not attempted were not included in either total. During the two-week intervention phase, students worked in pairs to practice the words and definitions for 5-10 minutes.

In Study 1, 18 general and special education teachers who taught grades 6-12 in resource, language arts, science, social studies, and math classes participated. Visual inspection of the graphed data as a group showed an increasing trend in correct rates during the intervention phase where over half of the group improved performance compared to baseline conditions. The typically achieving students (N = 236) improved as a group from a baseline mean accuracy score of 73% to an intervention score of 85%. Specifically, 69% improved, 11% stayed about the same, and 19% worsened in performance. By comparison, students with LD who received instruction in general education settings (N = 34) showed 72% improving, 8% about the same, and 20% worsening. Finally, students with LD in resource rooms (N = 57) demonstrated 49% improving, 23% remaining the same, and 28% worsening in performance. Retention accuracy scores remained comparable to the intervention phase scores for both groups.

In Study 2, the same 18 teachers implemented the intervention, which involved 34 students with LD. As a group, results showed a 21 percentage point change from baseline to intervention conditions (i.e., accuracy scores improved from 52% to 73% across conditions). Typically achieving students (N = 248) improved from 71% to 79% accuracy across baseline and intervention conditions, which represents an 8 percentage point difference. In terms of improvement in performance of the typically achieving students, 64% improved, 16% remained the same, and 20% worsened. Students with LD in general education settings (N = 30) demonstrated scores of 70% (improved), 10% (stable), and 20% (worsened) in performance, whereas students with LD in resource settings showed 80% improved, 0% stayed the same, and 20% worsened. Retention rates were slightly lower than Study 1 for both groups.

Mnemonic Strategy Instruction

Mnemonic strategy instruction is based on associative-memory facilitation that aids in recall and retention of information. Three articles were identified that studied the effects of mnemonics as an intervention for vocabulary acquisition (see Table 3).

Mastropieri et al. (1985) conducted two experiments to study the effects of the keyword mnemonic strategy on students' ability to recall the definitions of 14 vocabulary words. Taken from McGivern and Levin's (1983) materials, the words were described as low-frequency English vocabulary words. In Experiment 1, 32 school-identified students with LD participated. Students were randomly assigned to one of two conditions--mnemonic picture or direct instruction--and materials were developed for each condition. For the mnemonic picture condition, the vocabulary words, key words, and definitions were written on cards. The card depicted keywords, acoustically similar to the vocabulary words, interacting with the definitions (mnemonic pictures). In the direct instruction condition, only vocabulary words and their definitions were provided on index cards. One individually administered experimental session was conducted where students were provided the intervention to learn the vocabulary words. At the conclusion of the intervention, vocabulary words were presented in random order on a quiz that required students to provide the definitions. Results indicated that students in the mnemonic picture condition outperformed students in the direct instruction condition.

In Experiment 2, the investigators sought to understand the effects of student-generated interactive pictures (mnemonic imagery condition) compared to the same direct instruction procedures implemented in Experiment 1. In this study, vocabulary words, keywords, and definitions (no interactive mnemonic pictures) were again presented on index cards. Students were shown models of mnemonic interactive strategy pictures but were left on their own to construct interactive pictures for the vocabulary words. Again, the mnemonic imagery intervention was more effective as reflected in students' recall of vocabulary definitions than the direct instruction procedures for learning vocabulary meanings.

In another study, Condus, Marshall, and Miller (1986) examined the keyword mnemonic strategy as a means to improve vocabulary learning and retention. Sixty students (48 males and 12 females) with an identified learning disability in reading participated. Based on results of the Peabody Picture Vocabulary Test-Revised (PPVT-R; Dunn & Dunn, 1981), students were divided into two groups, high and low receptive vocabulary abilities, and randomly assigned to one of four conditions: keyword-image, picture context, sentence-experience context, and control.

Fifty vocabulary words were selected from the sixth-and eighth-grade Living Word Vocabulary (Dale & O'Rourke, 1981) curriculum and divided into five sets of 10 words, with two remaining words for practice items. Seven resource teachers taught the sets of 10 words across five weeks in 20-minute vocabulary lessons. For keyword-image instruction, vocabulary words were presented on index cards with the word on one side and the keyword and vocabulary word contained in a sentence on the reverse side. Pictorial line drawings showed representations of the keywords interacting with the definitions. In the picture context condition, students received cards similar to the keyword-image condition except that the illustrations did not contain keywords. A noun that had no acoustical similarity (stimulus recoding) to the vocabulary word was used instead of the keyword. Finally, for the sentence-experience context condition, students received three-sentence passages containing the vocabulary word.

During each week, three days of instruction were conducted and varied according to condition. Across the three interventions, students were introduced to the words according to the format designed for the study and then were given 20 seconds to study the words. On the third day for the keyword-image and picture context conditions, students were shown the 10 vocabulary words and asked to recall the keyword or picture, respectively. Corrective feedback was provided for no or incorrect responses. The resource teacher conducted the intervention or control condition in small groups. A multiple-choice test was conducted in phases, immediately after the weekly instruction and the completion of weekly tests (post), two weeks after the post-measure (maintenance), and eight weeks after maintenance testing (followup).

Results showed that overall, students in the keyword-image condition outperformed students in the other conditions under both immediate and long-term time intervals. Regarding receptive language, a significant main effect for receptive language ability (p = .0065) indicated that students with high receptive vocabularies outperformed students with low receptive vocabularies across all conditions both during immediate and weekly testing. However, at the eight-week followup testing, students with low receptive vocabulary abilities in the keyword-image condition outperformed students with low and high receptive vocabulary abilities in any of the other conditions, and student performance in the two contextual and control groups worsened over time.

In the third study in this category, Mastropieri, Scruggs, and Fulk (1990) examined the effects of the keyword method versus a rehearsal condition. Twenty-five middle school students received one-on-one instruction in a 16-minute experimental session. Two conditions were presented--the keyword and rehearsal treatments. The presentation of the vocabulary words was the main difference between the two conditions. In the keyword condition, vocabulary words, keywords, and definitions were presented on cards. The keywords were shown interacting with their definitions, whereas no pictorial representation was included in the rehearsal condition. Rather, the instructor focused on teaching procedures that included drill, practice, and corrective feedback. After a 1-minute activity, students were given production recall (post-intervention) and comprehension (generalization) tests. Results suggested that the keyword method facilitated recall and promoted generalization on a novel task.

Concept Enhancement Instruction

One article was identified that focused on the effects of concept enhancements as an intervention for vocabulary acquisition (see Table 4). Bos and Anders (1990) designed an intervention study that examined the effects of vocabulary instruction on vocabulary learning and reading comprehension. The study consisted of four conditions with an emphasis on learning and practicing definitions. A definition instruction (DI) condition was compared to activating student background knowledge, developing conceptual understanding, and building semantic relationships among word meanings through semantic mapping (SM), semantic feature analysis (SFA), or semantic/syntactic feature analysis (SSFA). Sixty-one junior high school students (41 male students, 20 female students) with an identified learning disability participated in the study. Instructional materials consisted of a practice and an experimental passage that were developed from science textbooks and trade books. Concept-related vocabulary was the instructional focus for the four conditions.

In the SM condition, using the list of vocabulary words, students generated a map that depicted a visual representation of the relationships among the vocabulary words. In the SFA condition, a relationship matrix was constructed by having the superordinate concept as the title. The vocabulary representing the coordinate concepts was placed along the top of the matrix and the vocabulary representing the subordinate concepts was placed along the side. In the SSFA condition, cloze-type sentences, which were developed based on the relationship matrix, were employed. Finally, in the DI condition, a list of vocabulary and their definitions was used.

The interventions occurred for eight 50-minute sessions over approximately seven weeks. Students and their assigned researcher participated in three 50-minute practice sessions. Then, students engaged in their assigned experimental condition across three days for 50-minute sessions. Across each condition, students were given 10 minutes to study with their instructional aide and 20 minutes to complete a recall test; additional time was given to complete a multiple-choice test (Day 3). Four weeks later a maintenance or followup session was conducted.

A 30-item multiple-choice test was administered post-intervention and as a maintenance check. Fifteen items focused on the context-related meanings of the vocabulary studied and 15 items made up a comprehension test measuring students' understanding of the reading and their ability to apply the vocabulary studied to new situations. Vocabulary and comprehension were measured with the multiple-choice test. A written recall test, where students wrote for 20 minutes on "all you know about the topic," was examined for vocabulary generated, conceptual units, and scriptal knowledge (use of prior knowledge). Students were also given a prior-knowledge test and topic interest inventory.

Overall, results revealed that at post-intervention testing, students who participated in the interactive, conceptually based interventions (SM, SFA, and SSFA) outperformed students who received definition instruction on vocabulary and comprehension items. Again at maintenance testing, students in the interactive conditions outperformed students in the DI condition on the 30-item vocabulary and comprehension measure and on the written recall for vocabulary generated, conceptual units, and scriptal knowledge.

DISCUSSION

The purpose of this article was to review the findings of vocabulary interventions designed for students with LD, and to present implications for vocabulary instruction to facilitate word-meaning learning and reading comprehension in content-area classes. Four categories of interventions were identified: computer-assisted instruction (CAI), fluency-building vocabulary practice activities, mnemonic strategy instruction, and concept enhancement instruction. Depending on the purpose of vocabulary instruction (e.g., conceptual understanding, semantic relationships), vocabulary interventions for teaching word meanings at the word-knowledge level and for processing knowledge at deeper levels of understanding to promote reading comprehension have been identified, albeit on a limited basis, for secondary students with LD. These interventions can be implemented prior to, during, and after reading to facilitate word-knowledge learning and reading comprehension.

Several findings are particularly noteworthy across the studies. All the interventions included in this review involved some type of manipulation, whether it was computer-assisted learning and interacting with text or creating associative and semantic relationships using illustrations, mnemonics, grids, and maps. Across studies, students showed gains within a relatively short duration of instruction; most of the studies provided regular instruction several times weekly, demonstrating that vocabulary instruction does not have to consume large amounts of class time to produce results. In all cases, vocabulary interventions were favored over vocabulary instruction that included more traditional instruction (e.g., dictionary usage, context clues) and instruction that included just explicit teaching (i.e., modeling how to find a definition, practice) (Bos & Anders, 1990; Condus et al., 1986; Mastropieri et al., 1985). Several of the studies paired mnemonic and direct instruction, suggesting that vocabulary instruction may be optimized when the features of both are incorporated (Mastropieri et al., 1990). For studies that included a transfer measure (i.e., reading comprehension), generalization effects were greater for interventions that involved deeper processing of word knowledge (i.e., mnemonics and concept enhancement instruction). This finding suggests that if the goal of vocabulary instruction is to teach word meaning, which can also transfer to reading comprehension, interactive interventions that feature associative learning, conceptual relationships, and prior knowledge should be selected (Bos & Anders, 1990). In conclusion, the results indicate that a variety of interventions have been documented as helpful in improving vocabulary knowledge that can transfer to promoting reading comprehension for students with LD across content-area subjects.

However, several limitations suggest caution in interpreting the results of this review. The fact that only six articles met the selection criteria suggests that research on effective vocabulary interventions for students with LD is limited, which is a significant problem given the importance of new word acquisition in content-area texts. This small number of articles also presents a narrow view of instructional options. Thus, other interventions that have proven effective for teaching students from a broader perspective (e.g., at-risk students, remedial students) should be explored as possible alternatives, as well as interventions that were discussed in this review.

Additionally, classroom teachers implemented only half of the studies. Given fidelity issues associated with teacher implementation of interventions (Jitendra & Xin, 1997), there is cause for concern about the extent of the effectiveness of the interventions. The conclusion is that future research is warranted in vocabulary interventions for secondary students with LD that consists of classroom teachers implementing the techniques and where fidelity of instruction is carefully monitored.

Finally, with the increasing number of technology tools that purport to assist with vocabulary development and enhancement (e.g., the Reading Pen, the visual thesaurus), research is necessary to determine the efficacy of these tools in helping students understand word meanings and comprehend text.

IMPLICATIONS

Even with the limitations of this review, several implications can be drawn. The most obvious is that use of independent word-learning strategies (i.e., using the dictionary and context clues) is insufficient for students with LD. These students possess limited vocabulary knowledge compared to their typically achieving peers who have spent more time engaged in wide reading, which promotes vocabulary development (Baker et al., 1995; Beck & McKeown, 1991). Therefore, depending on the desired depth of knowledge about key vocabulary, interventions that engage students interactively with memory devices (mnemonics) and graphic depictions (e.g., semantic maps, grids) and that are paired with direct instruction seem most promising in promoting word-meaning knowledge and reading comprehension of passages. Students also should have multiple exposures to words across time, including contextual exposures, to promote deeper understanding. Vocabulary instruction should be programmed into classroom teaching on a frequent basis and seems most beneficial when students work in small groups or in pairs to engage in vocabulary activities. Moreover, the number of words that students are expected to master at any given time should be kept relatively small. For example, the notion of 25 vocabulary words to be learned over the course of a week using only the definitional approach seems unlikely to produce the desired effect of teaching word knowledge for struggling students. For students who require additional practice beyond what teachers can reasonably provide, CAI is a promising tool that can be used independently by students for practice opportunities. Finally, although in need of further research, preliminary results suggest that vocabulary instruction that benefits students with LD helps typically achieving students as well. Therefore, the instructional time spent on vocabulary instruction can be worthwhile for all students in content-area classes.

Table 1
Computer-Assisted Interventions

Citation     Sample         Intervention

Johnson,     25 students;   Computer-Assisted
Gersten, &   grades 9-12;   Instruction Conditions:
Carnine      resource       Two CAI (1) (small vs. large
(1987)       room           teaching set) methods
                            focusing on the effects of
                            size of daily teaching sets
                            and provisions for
                            practice and review using
                            a multiple-choice,
                            recognition format on
                            time to mastery, posttest,
                            maintenance, and transfer
                            of vocabulary words;
                            4 times per week; 11 days
                            for 20 minutes each day.

                                            RESULTS

Citation     Sample         Post-Intervention       Maintenance

Johnson,     25 students;   Time to Mastery:        Slight drop in
Gersten, &   grades 9-12;   Mean number of          performance
Carnine      resource       sessions 7.6 for STS,   between posttest
(1987)       room           (2) 9.1 for LTS, (3)    and maintenance
                            stat. sign. p < .05     test--81% and
                            level; 10 (STS) vs. 8   84% for small
                            (LTS) students          and large
                            reached mastery; 83%    teaching set,
                            STS reached mastery     respectively.
                            compared to 67% LTS.
                            Posttest:
                            Mean performance--84%
                            and 87.8% for STS &
                            LTS, respectively; no
                            sign. effect for type
                            of instruction.
                            Attitude Survey:
                            Results favored STS.

Citation     Sample         Generalization

Johnson,     25 students;   For students in
Gersten, &   grades 9-12;   both teaching set
Carnine      resource       conditions, open-ended
(1987)       room           oral test of word
                            recognition about 35%
                            accuracy and test
                            of passage
                            comprehension
                            about 50%
                            accuracy.

(1) Computer-assisted instruction; (2) small teaching set; (3) large
teaching set.

Table 2
Fluency-Building Vocabulary Practice Sheets

Citation    Sample (1)            Intervention

Stump,      Study 1:              Study 1 & Study 2 (replication)
Lovitt,     GR 6-12, 236 gen.     Fluency-Building Practice
Fister,     ed., 91 students      Condition:
Kemp,       LD: 30 RR; 8 SS--GR   New vocabulary introduced
Moore, &    7; 20 SC--GR          by teachers, students studied
Schroeder   9, 10; 28 LA--GR      words and definitions
            6, 9, 10; 5           independently for 5-10 minutes;
            math--GR 7, 8         in pairs for 5-10 minutes,
                                  students quizzed each other
                                  on vocabulary words
                                  provided on practice sheets;
                                  students took turns supplying
                                  terms or definitions and
                                  checking responses using
                                  answer key; 3 days per week;
                                  10-20 minutes per day,
                                  2 weeks of intervention.

            Study 2:
            GR 7-11, 239 gen.
            ed., 34 students
            LD: 10 SS--GR
            7-9; 9 SC--GR 8,
            10, 11; 15
            LA--GR 7-9

                                            RESULTS

Citation    Sample (1)            Post-Intervention   Maintenance

Stump,      Study 1:              Study 1:            Study 1:
Lovitt,     GR 6-12, 236 gen.     Students with LD    For students with
Fister,     ed., 91 students      as a group          LD as a group,
Kemp,       LD: 30 RR; 8 SS--GR   improved from 68%   retention of
Moore, &    7; 20 SC--GR          correct during      baseline words
Schroeder   9, 10; 28 LA--GR      baseline to 80%     was 76% accuracy,
            6, 9, 10; 5           correct during      and 83% accuracy
            math--GR 7, 8         intervention.       for intervention
                                                      group (N = 78).

            Study 2:              Study 2:            Study 2:
            GR 7-11, 239 gen.     Students with LD    For students with
            ed., 34 students      as a group          LD as a group,
            LD: 10 SS--GR         improved from 52%   retention of
            7-9; 9 SC--GR 8,      correct during      baseline words
            10, 11; 15            baseline to 73%     was 60%, and 79%
            LA--GR 7-9            correct during      for intervention
                                  intervention.       words (N = 31).

Citation    Sample (1)            Generalization

Stump,      Study 1:              N/A (2)
Lovitt,     GR 6-12, 236 gen.
Fister,     ed., 91 students
Kemp,       LD: 30 RR; 8 SS--GR
Moore, &    7; 20 SC--GR
Schroeder   9, 10; 28 LA--GR
            6, 9, 10; 5
            math--GR 7, 8

            Study 2:
            GR 7-11, 239 gen.
            ed., 34 students
            LD: 10 SS--GR
            7-9; 9 SC--GR 8,
            10, 11; 15
            LA--GR 7-9

(1) GR = grade, RR = resource room, SS = social studies, SC = science,
LA = language arts; (2) Not Applicable.

Table 3
Mnemonic Strategy Instruction

Citation       Sample             Intervention

Mastropieri,   Study 1:           Study 1:
Scruggs,       32 students--21    Mnemonic Picture Condition: After two
Levin,         male, 11 female;   practice examples (modeling),
Gaffney, &     mean age--13       vocabulary words with mnemonic
McLoone        years 11 months;   pictures (keyword interacting with
(1985)         students in        definition) were presented.
               grades 7           Directions: "The keyword for ranid is
               (N = 10), 8        rain (stimulus recoding). Ranid means
               (N = 11), and 9    frog (semantic relating). Remember
               (N = 11); CA       this picture of a frog sitting in the
               Ach. Test mean     rain. Remember this picture of what?
               reading            And ranid means what?" (systematic
               score--24.9;       retrieving) (direct instruction
               study conducted    paired with mnemonic strategy).
               in separate room   Students provided definitions and
               in the school      described the pictures.
                                  Direct Instruction Condition:
                                  Students told the definition, which
                                  they repeated: "ranid means frog,
                                  what does ranid mean?" (direct
                                  instruction). The words were taught
                                  in sets of five and five, then all
                                  10 words were reviewed, and in a set
                                  of four followed by all 14 words
                                  reviewed using direct instruction. 1
                                  session and approximately 10-15
                                  minutes--mnemonic pictures; 20
                                  minutes--direct instruction.

                                  Study 2:
               Study 2:           Mnemonic Imagery Condition: Two
               30 students--23    practice items were shown: "The word
               male, 7 female;    clue for ranid is rain (stimulus
               mean age--13       recoding), and ranid means frog
               years 10 months;   (semantic relating). Think of a
               students in        picture of rain and a frog doing
               grades 7           something together;" interactive
               (N = 14), 8        pictures were shown as examples
               (N = 8), and 9     (modeling). Students made own
               (N = 8); CA (1)    interactive pictures
               mean reading       (student-generated) for vocabulary
               score--27.6;       words followed by seeing sample
               study conducted    drawings (feedback).
               in separate room   Direct Instruction Condition:
                                  Procedures similar to mnemonic
                                  imagery condition, except no keywords
                                  or pictured references to keywords.

                                              RESULTS

Citation       Sample             Post-Intervention   Maintenance

Mastropieri,   Study 1:           Study 1:            Study 1:
Scruggs,       32 students--21    Mean recall of      N/A (2)
Levin,         male, 11 female;   mnemonic-picture
Gaffney, &     mean age--13       students (79.5%)
McLoone        years 11 months;   was substantially
(1985)         students in        and statistically
               grades 7           higher than that
               (N = 10), 8        of the direct
               (N = 11), and 9    instruction only
               (N = 11); CA       students (31.2%),
               Ach. Test mean     t(30) = 7.12,
               reading            p < .001.
               score--24.9;
               study conducted
               in separate room
               in the school

               Study 2:           Study 2:            Study 2:
               30 students--23    Mean definition     N/A
               male, 7 female;    recall of
               mean age--13       mnemonic-imagery
               years 10 months;   students (69.3%)
               students in        was statistically
               grades 7           higher than that
               (N = 14), 8        of direct
               (N = 8), and 9     instruction only
               (N = 8); CA (1)    subjects (46.7%),
               mean reading       t(28) = 2.96,
               score--27.6;       p < .01.
               study conducted
               in separate room

Citation       Sample             Generalization

Mastropieri,   Study 1:           Study 1:
Scruggs,       32 students--21    N/A
Levin,         male, 11 female;
Gaffney, &     mean age--13
McLoone        years 11 months;
(1985)         students in
               grades 7
               (N = 10), 8
               (N = 11), and 9
               (N = 11); CA
               Ach. Test mean
               reading
               score--24.9;
               study conducted
               in separate room
               in the school

               Study 2:           Study 2:
               30 students--23    N/A
               male, 7 female;
               mean age--13
               years 10 months;
               students in
               grades 7
               (N = 14), 8
               (N = 8), and 9
               (N = 8); CA (1)
               mean reading
               score--27.6;
               study conducted
               in separate room

(1) CA = California Achievement Test; (2) Not Applicable.

Citation       Sample             Intervention

Condus,        60 students--48    Keyword-Image Condition: Students
Marshall, &    males, 12          asked to use interactive pictures to
Miller         females; mean      remember vocabulary word meanings
(1986)         age--12 years 6    (keyword mnemonics). Students learned
               months; PPVT-R     keywords, then supplied keywords when
               SS (1) 85 for      vocabulary words presented. After 20
               high and 70 SS     seconds of study time, students read
               for low            content aloud, then stated keyword
               receptive          and illustration for each vocabulary
               language; study    word (practice).
               conducted in       Picture Context Condition: The
               junior high        teacher read each word, definition,
               resource room      and two-sentence conversation for the
                                  picture (modeling). Students studied
                                  pictures for 20 seconds, then read
                                  content to the teacher (practice).
                                  Sentence-Experience Context
                                  Condition: The teacher read sentences
                                  for the words. Students were given a
                                  question to answer and asked to
                                  relate the meaning of the word to
                                  their experiences (practice).
                                  Students given 20 seconds additional
                                  study time.
                                  Control Group Condition: The teacher
                                  read words and meanings. Students
                                  chose own way to learn the vocabulary
                                  meanings (independent practice). On
                                  subsequent study days, the teacher
                                  said correct pronunciation of the
                                  word if students had difficulty
                                  reading words; 20 minutes of training
                                  3 days per week for 5 weeks.

Mastropieri,   25 students--17    Keyword Condition: Students learned
Scruggs, &     males, 8           keywords for 16 vocabulary words.
Fulk (1990)    females; mean      Students saw each interactive picture
               age--14 years 1    for 30 seconds while experimenter
               month; students    described keyword and pictorial
               grade 6 (N = 5),   strategy (mnemonic strategy).
               7 (N = 9), and     Students provided definition of each
               8 (N = 11);        vocabulary word and described the
               mean WRAT-R GE     picture (practice).
               (2) for            Rehearsal Condition: After practice
               reading--5.0;      items, students pronounced vocabulary
               study conducted    words and learned words through drill
               in separate room   and practice, rapid-paced
                                  questioning, and corrective feedback;
                                  1 session for 16 minutes.

                                                 RESULTS

Citation       Sample             Post-Intervention   Maintenance

Condus,        60 students--48    Immediate and       Maintenance (2
Marshall, &    males, 12          weekly tests:       weeks later) and
Miller         females; mean      Students in all     followup (8 weeks
(1986)         age--12 years 6    treatment           later): Keyword
               months; PPVT-R     conditions          students
               SS (1) 85 for      recalled high       significantly
               high and 70 SS     percentage of       outperformed
               for low            words on            students in other
               receptive          immediate test.     conditions both
               language; study    Keyword students    times. Students
               conducted in       significantly       with high
               junior high        outperformed        receptive
               resource room      students in other   vocabularies
                                  treatment           outperformed
                                  conditions on the   students with low
                                  weekly tests. All   receptive
                                  treatment groups    vocabularies in
                                  significantly       both maintenance
                                  outperformed        conditions.
                                  students in the     Students with low
                                  control             receptive
                                  condition.          vocabularies in
                                  Students with       the keyword-image
                                  high receptive      condition
                                  vocabularies        performed better
                                  performed better    than students
                                  than students       with high and low
                                  with low            receptive
                                  receptive           vocabularies in
                                  vocabularies.       the other
                                                      conditions at the
                                                      8-week followup
                                                      to testing.

Mastropieri,   25 students--17    On the production   N/A
Scruggs, &     males, 8           test, students in
Fulk (1990)    females; mean      the keyword
               age--14 years 1    mnemonic
               month; students    condition
               grade 6 (N = 5),   outperformed
               7 (N = 9), and     students in the
               8 (N = 11);        rehearsal
               mean WRAT-R GE     condition. A main
               (2) for            effect was found
               reading--5.0;      for item type on
               study conducted    the production
               in separate room   test, with
                                  students
                                  performing better
                                  on concrete
                                  words.

Citation       Generalization

Condus,        60 students--48    N/A
Marshall, &    males, 12
Miller         females; mean
(1986)         age--12 years 6
               months; PPVT-R
               SS (1) 85 for
               high and 70 SS
               for low
               receptive
               language; study
               conducted in
               junior high
               resource room

Mastropieri,   25 students--17    On the comprehension
Scruggs, &     males, 8           test, the keyword
Fulk (1990)    females; mean      mnemonic condition
               age--14 years 1    surpassed the
               month; students    rehearsal condition.
               grade 6 (N = 5),   No main effect for
               7 (N = 9), and     item type was found
               8 (N = 11);        on the comprehension
               mean WRAT-R GE     test.
               (2) for
               reading--5.0;
               study conducted
               in separate room

(1) PPVT--R SS = Peabody Picture Vocabulary Test--Revised Standard
Score; (2) WRAT--R = Wide Range Achievement Test--Revised, GE = Grade
Equivalent.

Table 4
Concept Enhancement Instruction

Citation       Sample             Intervention

Bos & Anders   61 students:       Semantic Map (SM) Condition:
(1990)         41 males,          Prior to reading, students
               20 females; mean   participated in interactive
               age--13.80;        discussion about topic; with
               junior high        researcher, developed a hierarchical
               grade; Woodcock    relationships map using selected
               Reading Mastery    vocabulary words (Day 1); after
               Test--Revised      reading, discussed their maps and
               mean reading       confirmed definitions and/or
               score 81.30;       relationships by using text (Day 2).
               study conducted    Semantic Feature Analysis (SFA)
               in large group     Condition:
                                  Students completed prereading
                                  activity (relationship matrix) to
                                  activate prior knowledge, predict and
                                  draw relationships among concepts
                                  (Day 1); purpose for reading was
                                  discussed, students read the passage.
                                  After reading, researcher and
                                  students discussed matrix, Students
                                  used text to confirm definitions
                                  and/or relationships (Day 2).
                                  Semantic/Syntactic Feature Analysis
                                  (SSFA) Condition:
                                  Students engaged in interactive
                                  activity to activate prior knowledge,
                                  predict and draw relationships among
                                  concepts (Day 1); students read
                                  passage, then discussed matrix and
                                  sentences. Students used text to
                                  confirm information (Day 2).
                                  Definition Instruction (DI)
                                  Condition:
                                  Students taught definitions of
                                  vocabulary words and instructed to
                                  memorize definitions (Day 1);
                                  students read passage and reviewed
                                  vocabulary and definitions. Students
                                  used text to confirm definitions
                                  (Day 2); 50-minute experimental
                                  sessions.

                                                 RESULTS

Citation       Sample             Post-Intervention   Maintenance

Bos & Anders   61 students:       Students in the     Students in the
(1990)         41 males,          SFA and SM          three interactive
               20 females; mean   SFA outperformed    intervention
               age--13.80;        students in the     groups
               junior high        definition          outperformed
               grade; Woodcock    condition; no       students in the
               Reading Mastery    significant         direct
               Test--Revised      differences among   instruction
               mean reading       the interactive     group.
               score 81.30;       conditions.
               study conducted
               in large group

Citation       Sample             Generalization

Bos & Anders   61 students:       Post-Intervention:
(1990)         41 males,          Comprehension--the main-effect test
               20 females; mean   results for condition and time were
               age--13.80;        significant (p <. 001). Post-hoc
               junior high        analyses showed that students in the
               grade; Woodcock    interactive conditions scored
               Reading Mastery    significantly higher than students in
               Test--Revised      the DI condition; no differences
               mean reading       among the three interactive
               score 81.30;       conditions.
               study conducted    Conceptual Units--Post-hoc analyses
               in large group     showed no significant difference
                                  among the four instructional
                                  conditions.
                                  Scriptal Knowledge--Post-hoc analyses
                                  for the adjusted means indicated no
                                  differences among groups.

                                  Maintenance:
                                  Comprehension--Students in the SSFA
                                  condition outperformed those in the
                                  DI condition; no differences among
                                  the three interactive conditions.
                                  Vocabulary Generated--Students in the
                                  DI condition generated significantly
                                  less vocabulary than students in
                                  either the SSFA or SFA conditions.
                                  Conceptual Units--Students in the
                                  SSFA and SFA condition generated
                                  significantly more conceptual units
                                  than students in the DI condition.
                                  There were no differences among the
                                  three interactive conditions.
                                  Scriptal Knowledge--Students in the
                                  three interactive conditions
                                  generated higher quality in their
                                  recalls than students in the DI
                                  condition.

Requests for reprints should be addressed to: Diane Pedrotty Bryant, Dept. of Special Education, D5300, University of Texas at Austin, Austin, TX 78712.

REFERENCES

Baker, S. K., Simmons, D. C., & Kameenui, E.J. (1995). Vocabulary acquisition: Curricular and instructional implications for diverse learners (Technical Report No. 14). Eugene: University of Oregon, National Center to Improve the Tools of Educators.

Beck, I., & McKeown, M. (1991). Conditions of vocabulary acquisition. In R. Barr, M. L. Kamil, P. Mosenthal, & P. D. Pearson (Eds.), Handbook of reading research (Vol. 2, pp. 789-814). New York: Longman.

Bos, C. S., & Anders, P. L. (1990). Effects of interactive vocabulary instruction on the vocabulary learning and reading comprehension of junior-high learning disabled students. Learning Disability Quarterly, 13, 31-42.

California Achievement Tests, Fifth Edition. (1992). Monterey, CA: CTB/McGraw-Hill.

Carnine, D., Kameenui, E. J., & Coyle, G. (1984). Utilization of contextual information in determining the meaning of unfamiliar words. Reading Research Quarterly, 19, 188-204.

Condus, M. M., Marshall, K. J., & Miller, S. R. (1986). Effects of the keyword mnemonic strategy on vocabulary acquisition and maintenance by learning disabled children. Journal of Learning Disabilities, 19(10), 609-613.

Dale, E., & O'Rourke, J. (1981). Living word vocabulary. Chicago: Childcraft.

Dunn, L. M., & Dunn, L. M. (1981). Peabody Picture Vocabulary Test--Revised. Circle Pines, MN: American Guidance Service.

Jastak, S., & Wilkinson, G. S. (1984). Wide Range Achievement Test Revised. Wilmington, DE: Jastak Associates.

Jitendra, A., & Xin, Y. P. (1997). Mathematical word-problem-solving instruction for students with mild disabilities and students at risk for math failure: A research synthesis. The Journal of Special Education, 30(4), 412-438.

Johnson, G., Gersten, R., & Carnine, D. (1987). Effects of instructional design variables on vocabulary acquisition of LD students: A study of computer-assisted instruction. Journal of Learning Disabilities, 20(4), 206-213.

Margosein, C. M., Pascarella, E. T., & Pflaum, S. W. (1982, April). The effects of instruction using semantic mapping on vocabulary and comprehension. Paper presented at the annual meeting of the American Educational Research Association, New York.

Mastropieri, M. A., Scruggs, T. E., & Fulk, B. J. (1990). Teaching abstract vocabulary with the keyword method: Effects on recall and comprehension. Journal of Learning Disabilities, 23(2), 92-107.

Mastropieri, M. A., Scruggs, T. E., Levin, J. R., Gaffney, J., & McLoone, B. (1985). Mnemonic vocabulary instruction for learning disabled students. Learning Disability Quarterly, 8, 57-63.

McGivern, J. E., & Levin, J. R. (1983). The keyword method and children's vocabulary learning: An interaction with vocabulary knowledge. Contemporary Educational Psychology, 8, 46-54.

McKeown, M. G. (1993). Creating effective definitions for young word learners. Reading Research Quarterly, 28, 16-31.

Nagy, W. (1988). Teaching vocabulary to improve reading comprehension. Newark, DE: International Reading Association.

Nagy, W., & Stahl, S. (2000). Promoting vocabulary development. Austin: Texas Education Agency.

Patberg, J. A., & Stibbe, M. A. (1985, December). The effects of contextual analysis instruction on vocabulary learning. Paper presented at the National Reading Conference, San Diego, CA.

Pittelman, S. D., Levin, K. M., & Johnson, D. D. (1985). An investigation of two instructional settings in the use of semantic mapping with poor readers (Program Report No. 85-4). Madison: University of Wisconsin-Madison, Wisconsin Center for Education Research.

Pressley, M., & Woloshyn, V. (1995). Cognitive strategy instruction that really improves children's academic performance (2nd ed.). Cambridge, MA: Brookline Books.

Simmons, D. C., & Kameenui, E. J. (1990). The effect of task alternatives on vocabulary knowledge: A comparison of students with learning disabilities and students of normal achievement. Journal of Learning Disabilities, 23(5), 291-297.

Stahl, S. A. (1986). Three principles of effective vocabulary instruction. Journal of Reading, 662-668.

Stahl, S. A., & Fairbanks, M. M. (1986). The effects of vocabulary instruction: A model-based meta-analysis. Review of Educational Research, 56, 72-110.

Stump, C. S., Lovitt, T. C., Fister, S., Kemp, K., Moore, R., & Schroeder, B. (1992). Vocabulary intervention for secondary-level youth. Learning Disability Quarterly, 15, 207-222.

DIANE PEDROTTY BRYANT, Ph.D., is associate dean/associate professor, University of Texas.

MARILYN GOODWIN, Ph.D., is assistant professor, Southwest Texas State University.

BRIAN R. BRYANT, Ph.D., is president, Psycho-Educational Services, Austin, Texas.

KELLIE HIGGINS is a doctoral student, University of Texas.