In this report, we reviewed how reading components are related to literacy levels, focusing on Below Level 1 and Level 1 proficiency. We described the rationale and theoretical foundations for including reading components in the PIAAC survey. We chose a sample of English-speaking countries (United States, Canada, United Kingdom, and Ireland) so that we could compare U.S. performance to other countries where adults learn to read in English. We also included Spain and Italy, because the sight-to-sound systems of Spanish and Italian are more transparent than in English, potentially making it easier to learn foundational skills in these languages than in English. On the other hand, these countries scored near the bottom of the cross-country literacy comparison internationally. Key results concerning the United States adults and six-country sample included the following:
- General: The U.S. Below Level 1 and at Level 1 groups consistently scored below the other English-speaking countries in the sample, as well as below Spain and Italy. In most analyses, by Level 2, the U.S. results were comparable to other countries. All countries' results were near asymptote levels on reading components by Level 3.
- Print Vocabulary: The U.S. average percentage correct for print vocabulary lagged behind the six-country average for Below Level 1 (77% vs. 86%) and Level 1 (89% vs. 93%) before showing comparably high performance at Level 2 (95% vs. 97%). On the positive side, the performance was well above chance levels, suggesting that even Below Level 1 U.S. adults could recognize printed forms of words associated with common objects such as animals, furniture, and shapes.
- Sentence Processing: In the sentence processing task set, the United States lagged even further behind the six-country average for Below Level 1 (52% vs. 69%), Level 1 (72% vs. 81%), and Level 2 (87% vs. 91%) before reaching parity at Level 3 (95% for both). With chance performance at about 50%, this indicates that many Below Level 1 adults in the United States were not able to understand and evaluate even short sentences.
- Passage Comprehension: In the passage comprehension task set, the United States again lagged far behind the six-country average for Below Level 1 (47% vs. 62%) and at Level 1 (73% vs. 81%), before reaching parity at Level 2 (90% vs. 92%). With chance performance at about 50%, this indicates that many Below Level 1 adults in the United States were not able to read passages for basic understanding. In general, Below Level 1 adults performed relatively lower on the passage comprehension task set than on the sentence processing, but by Level 1, mean performance levels were comparable between these two reading component task sets.
- Reading Components Rate: With respect to time to complete the task sets, the United States was not disproportionately slower than the other countries in the sample. In general, differences in accuracy of performance were all reflected in speed or rate for completing task sets across the entire ability distribution. This suggests that fluency or automaticity of component skill processing is part of the underlying foundation of literacy for most adults. That is, Below Level 1 adults needed more time to achieve their levels of performance than Level 1 adults, who themselves needed more time than Level 2 adults, and so on across Levels 3, 4, and 5.
- Nonnative Speakers of the Test Language: A significant proportion of the difference between the United States and the other countries in the six-country sample could be attributed to the relatively poorer performance of the nonnative English-speaking subgroup in the U.S. sample. For the United States alone, nonnative speakers had lower mean performance scores than native English speakers across the summed total of the three component tasks at Below Level 1 (47% vs. 76%), Level 1 (65% vs. 86%), Level 2 (79% vs. 95%), and even at Level 3 (92% vs. 98%), before reaching parity at Level 4/5 (99% for both groups). The nonnative speakers in the U.S. sample generally performed lower than nonnative speaker subpopulations in other countries in the sample. In general, nonnative speaker groups in countries scored lower than native speakers. This pattern of results is consistent with the interpretation that performance on reading components tasks may be negatively impacted by weakness in individuals' language ability in the test language. The overall impact on national means would vary as a function of the relative size of the nonnative-speaking subpopulation in each country.
- Native Speakers of Test Language only: Comparing only native-speaking adults across the sample, the U.S. results were comparable to the six-country averages. For percentage correct scores (summed across the three task sets), the United States versus six-country sample results were: Below Level 1 (76% vs. 78%), Level 1 (86% vs. 89%), and Level 2 (95% for both). In the United States, native speakers of the test language who were Below Level 1 had performance levels of 92%, 72%, and 66% for print vocabulary, sentence processing, and passage comprehension, respectively. Although the United States native-speaking English sample scored comparable to the other sampled countries, this pattern still demonstrated weaknesses in reading component skills for at or below Level 1 U.S. adults (as well as in the other countries).
- Computer versus No Computer Experience: In general, there was a mean difference between adults who reported some computer experience versus those reporting no computer experience, but this mean difference diminished by Level 2 or 3 literacy proficiency level in most other countries in the sample. By contrast, the mean difference based on computer experience in U.S. adults persisted across proficiency levels. Note that this result was based entirely on the paper-based pathway subsample in which adults were administered the main assessments and reading components in paper booklets. That is, computer experience was associated with higher literacy scores on the paper-administered instruments.
- Associations among Reading Components and Literacy Proficiency Scores: Correlations among variables showed a small to moderate association between reading component accuracy and literacy proficiency, with a relatively larger association for sentences and passages than vocabulary. A negative correlation was found for the time to complete reading component task sets, indicating that as proficiency increased, the time to complete the tasks decreased.
- Multiple Regression Models: These models demonstrated the theoretically predicted relationship between reading component accuracy and rate with literacy scores. The model for the international average showed a small amount of variance (4%) predicted by print vocabulary alone, more with sentence processing (13%) or passage comprehension (14%) alone, and 17% when all three were used together. Adding the time to complete information resulted in a stronger relationship (30%) across the entire ability distribution, confirming the association of efficient (accuracy plus rate) text processing as skill levels increase. The relationship was relatively stronger in the U.S. sample in comparison to the six countries sampled.
Discussion and Implications for Policy and Practice
In this section, I first contextualize the results of this study as they relate to broad policy considerations. This is followed by some more specific instructional recommendations for developing foundational skills in adults, drawn primarily from the reading and learning science literature.
While developing advanced reading literacy proficiency requires practice applying skills in real-world settings, basic skill development additionally may require direct skill instruction and practice applying nascent skills in exercises that build up fluency of application of those skills. Consequently, understanding adults' performance on foundational skill tasks, such as those represented in the reading component task sets, provides different types of information than is gathered from adults' patterns of responses to applied reading literacy tasks, as represented in the main literacy survey tasks. By assessing foundational skills directly, the reading components results provide new information for practice and policy beyond general reading proficiency level scores. One can draw some implications for instructional programs from these results when coupled with the research literature on teaching adult learners.What policy implications can we draw from the basic science and empirical evidence of effective instruction for helping adults who still demonstrate foundational reading skill weaknesses to become better readers?
The good news that stems from the PIAAC reading components battery results is that adults in the United States who scored at literacy proficiency Level 3 or above (and the vast majority at Level 2) demonstrated near-ceiling level performances on reading component task sets, suggesting well-developed foundational reading skills. There was also some cause for optimism in the generally competent performance of many U.S. adults who were at or below Level 1, at least for native English-speaking U.S. adults. The performance of this group in recognizing high-frequency print vocabulary and evaluating the meaning of simple sentences demonstrated some basic literacy skills. The results, however, also revealed continued weaknesses in foundational skills for adults scoring below Level 1, and for many scoring at Level 1, in accuracy, but especially in fluency and automaticity of processing.
The reading components results are consistent with the claim that fluent, efficient basic skills are foundational in supporting development of more advanced literacy skills.49 Learning, especially formal learning that may occur in education or training programs, often includes learning new content and skills through evidence-based instruction. Empirical studies in the reading and learning sciences over the past several decades have yielded a rich literature for understanding reading processes,50 sound pedagogical approaches to learning to read,51 and descriptions and insights into dyslexia and reading disability, when development goes awry.52 The research is richest for young children but substantial in describing and seeking to understand those who struggle to read in adolescence and adulthood, as documented in a recent National Academies of Research Committee report.53 What implications can be drawn from this literature?
For one, we can conclude that it is unlikely that it will be easier and quicker for low-skill adults to learn to read fluently and well compared to what one would expect of children learning to read. The small number of studies that have attempted to accelerate adult reading progress via well-designed, research-based intervention programs have reported significant but modest gains,54 which is similar to results obtained with adolescent, in-school struggling readers.55 Direct instructional interventions of 200 hours or less have not been found sufficient for helping adults at levels equivalent to Below Level 1 or Level 1 to achieve high levels of literacy.56 Thus, targeted instruction and practice is effective but not a quick fix. This has policy implications for the expectations set for literacy programs with respect to learner achievement gains and the provision of learning and instructional duration and intensity.
Specifically, stakeholders need to be realistic about the progress one can expect in the short term from instructional programs. Further, policies need to be more sensitive to what it means to be an adult lifelong learner. It seems unlikely that thousands of hours of classroom instruction is a feasible policy alternative for delivering instruction to adults with skills at or below Level 1. Nor is it likely that adults would participate in such programs. Research shows that adult learners often have intermittent periods of self-study and program participation across the life span.57 In contrast, the interconnected, digital literacy world could bring services and tailored support to adults across intermittent periods of self-study and program participation. The promise of this type of adaptive instruction has been anticipated for decades,58 but perhaps now the circumstances are right for this promise to be realized or at least revisited. Digital technologies could provide both instructional support, remotely delivered, and access to the thousands of hours of immersion with text that may be necessary for adults to achieve higher reading levels. But for such a recommendation to be feasible, both access and readiness to use digital literacy tools needs to be established.
In addition, before we draw more dour conclusions about the instructional and the provision of time that low proficiency adults may require to achieve high levels of literacy proficiency, perhaps it is worth rethinking what is required (and provided) for children to acquire literacy proficiency. U.S. primary school children that are on a developmental trajectory that would eventually lead to college and career readiness typically still require two to three years in an immersive, daily reading literacy instructional and practice environment (that is, grades 1 to 3 of primary school) before becoming relatively fluent readers of simple texts. By grade 4, these normally developing learners (i.e., at the 50th percentile or above on a reading fluency measure) are reading fourth-grade level texts aloud at a rate of about 90 to 120 words per minute.59 Most fourth-grade level texts employ high frequency words and simple, short sentences. They are less complex in vocabulary, sentence, and discourse structure than middle-grades expository text that are used to teach subject domains such as history or science.60 Nonetheless, with frequent exposure and practice reading a wide variety of text structures and genres in each successive year (that is, grades 4-12), students on this fluency trajectory can expect to see their reading rate increase to average about 250 words per minute during silent reading of average adult-level prose texts.61This typical 12-year, immersive, developmental learning program may seem infeasible for adult education, but adult learners have their entire lives to traverse this learning space. Policy formulations should draw upon the reading and learning sciences literature, take advantage of digital technologies, and be more sensitive to what it means to be a lifelong learner.
The results of such formulations may challenge the assumptions of current infrastructure and funding priorities of the adult education field as it exists today, with perhaps a stronger emphasis on digital technologies. Toward this end we note again, however, that U.S. adults who reported no computer experience scored lower on the reading components, even though those reading components tasks were administered as a paper-based instrument. Given that so much personal, workplace, and commercial reading literacy takes place on or is migrating to electronic devices (e.g., smartphones, tablets, personal computers), and the promise and potential of digital literacy as part of the policy solution for adult education provision, the PIAAC results concerning computer experience of U.S. adults is a most troubling finding. Further research needs to be conducted to understand the source of this difference. Reder, for example, in an examination of the general PIAAC survey results, discussed pathways toward digital literacy and equity issues with respect to access and readiness to achieve high levels of digital literacy.62 He concluded that inequities exist based on race/ethnicity and national origin—a problem that policies could be designed to confront.
This is an appropriate segue to another central result revealed in the reading components analyses. A significant proportion of low-scoring U.S. adults were also nonnative speakers of English. Research, policies, and literacy programs will continue to need to prioritize understanding and serving the learning needs of nonnative English speakers in the United States. The current Workforce Innovation and Opportunity Act of 201463 does stipulate policies to fund programs for integrated English literacy and civics education including reading, writing, speaking, and comprehension skills in English. The law specifically refers to the term ‘‘essential components of reading instruction" as used in the Elementary and Secondary Education Act.64 It further goes on to encourage the development and use of promising assessment tools and strategies that are based on scientifically valid research to identify the needs of nonnative-speaking students at the lowest achievement levels. The evidence reported here confirms the need for further attention and research on effective learning strategies and assessments for nonnative speakers of English in the United States.
What instructional implications might be drawn for adults with weak foundational reading skills?
For adults who continue to experience challenges at the word level (including not only the breadth and depth of their vocabulary, but also fluent and automatic decoding and word recognition), adapting methods used when teaching beginning readers (or analogous methods for English language learners) may be appropriate.65 For example, Greenberg et al.66 report deficits in adult learner vocabulary relative to grade-matched children, suggesting that weaknesses in word recognition and learning skills may impede normal development of the breadth and depth of adult vocabulary.67 A growing body of research also suggests that morphological awareness, and morphology knowledge and skills more generally, is related to reading comprehension, as well as the subskills that underlie reading.68
For adults who continue to experience challenges at the sentence level, the root causes could include both inefficiencies at the word level as noted above and other semantic and syntactic processes. Complex sentences often have multiple embedded phrases and clauses that increase the distance between subjects and predicates, a feature known to increase processing demands.69 Key to understanding complex sentences, and perhaps instructionally malleable, is efficient processing of connectors. Relationships that are signaled in connectors may be temporal (e.g., before), causal (e.g., because), adversative (e.g., although), or conditional (e.g., if). Empirical studies have been conducted examining the difficulties learners often have in adequately processing these kinds of semantic70 or syntactic relations.71
For adults who continue to experience challenges at the passage comprehension level, a strong instructional program might involve varying the linguistic complexity of texts that the individual reads and providing opportunities for discussion of texts of increasing length and complexity.72 Continuous texts can also include the range of meaning relationships that are represented in the sentence processing items—referential, causal, and knowledge-based relationships among discourse entities.73 Adult learners can discuss their understanding of texts or try to summarize or explain it to others, whether to themselves, a teacher, or other adult learners.74 This interactive discussion may reveal errors in interpretation and can help readers build awareness and fluency in encoding sentence meaning for increasingly complex sentence and text structures.75
To build adults' reading fluency and stamina, literacy instruction should include opportunities for frequent practice with reading. Repeated reading of familiar texts is a technique that has been shown to help build reading fluency.76 Proficient readers typically read widely and have the stamina and perseverance to read lengthy texts (e.g., novels) yet are still able to retain in memory a general mental model of what they have read. The passage tasks model that kind of continuous reading practice and therefore serve as indicators of an adult's readiness for learning and instruction with longer, continuous texts.
Finally, helping low-skill adults to increase their reading proficiency will likely benefit from contextualized instruction, taking advantage of the broader knowledge of the world and relevant interests of the adult learners. Isolating all instruction on components is not likely to be as effective as finding the appropriate mix for each adult.77
Though not directly addressed in PIAAC, research on individuals with learning, mental, or physical disabilities should be considered when interpreting the results here, as it is likely that the scores of some segment of the sampled individuals may have been impacted by preexisting learning, mental, or physical conditions, with consequences for the instruction or policies that will be effective in helping them to enhance their reading abilities.78
49 National Research Council, Improving Adult Literacy.
50 Abadzi, Improving Adult Literacy Outcomes; Baer et al., Basic Reading Skills; Bell and Perfetti, Reading Skill; Cunningham, Stanovich, and Wilson, Cognitive Variation; Curtis, Development of Components; Perfetti, Universal Grammar; Sabatini, "Efficiency in Word Reading"; Sabatini, Learning and Assessment; Sabatini and Bruce, PIAAC Reading Components; Stine-Morrow, Miller, and Hertzog, Language Processing.
51 National Institute of Child Health and Human Development, Teaching Children to Read; National Research Council, Improving Adult Literacy; Rayner, Psychological Science.
52 National Institute for Literacy, Developing Early Literacy (Report of the National Early Literacy Panel) (Washington, DC: Author, 2008), http://lincs.ed.gov/publications/pdf/NELPReport09.pdf.
53 National Research Council, Improving Adult Literacy.
54 Judith A. Alamprese, Charles A. MacArthur, Cristofer Price, and Deborah Knight, "Effects of a Structured Decoding Curriculum on Adult Literacy Learners' Reading Development," Journal of Research on Educational Effectiveness 4, no. 2 (2011): 154-72; Anderson and Freebody, "Vocabulary Knowledge"; Daphne Greenberg, Justin C. Wise, Robin Morris, Laura D. Fredrick, Victoria Rodrigo, Alice O. Nanda, and Hye K. Pae, "A Randomized Control Study of Instructional Approaches for Struggling Adult Readers," Journal of Research on Educational Effectiveness 4, no. 2 (2011): 101-17; Michael F. Hock and Daryl F. Mellard, "Efficacy of Learning Strategies Instruction in Adult Education," Journal of Research on Educational Effectiveness 4, no. 2 (2011): 134-53; John P. Sabatini, Jane Shore, Steven Holtzman, and Hollis S. Scarborough, "Relative Effectiveness of Reading Intervention Programs for Adults with Low Literacy," Journal of Research on Educational Effectiveness 4, no. 2 (2011): 118-33.
55 Sharon Vaughn, Elizabeth A. Swanson, and Michael Solis, "Reading Comprehension for Adolescents with Significant Reading Problems," in Handbook of Learning Disabilities (2nd Ed.), ed. H. Lee Swanson, Karen R. Harris and Steve Graham (New York: Guilford Press, 2013), 375-87.
56 National Research Council, Improving Adult Literacy.
57 Stephen Reder and John Bynner, eds., Tracking Adult Literacy and Numeracy Skills: Findings from Longitudinal Research (New York and London: Routledge, 2009).
58 For example, Office of Technology Assessment, Adult Literacy and New Technologies: Tools for a Lifetime (Washington, DC: U. S. Government Printing Office, 1993).
59 Mary C. Daane, Jay R. Campbell, Wendy S. Grigg, Madeline J. Goodman, and Andreas Oranje, Fourth-Grade Students Reading Aloud: NAEP 2002 Special Study of Oral Reading (Washington, DC: U.S. Department of Education. Institution of Education Sciences, National Center for Educational Statistics, 2005).
60 Paul Deane, Kathleen M. Sheehan, John Sabatini, Yoko Futagi, and Irene Kostin, "Differences in Text Structure and Its Implications for Assessment of Struggling Readers," Scientific Studies of Reading 10, no. 3 (2006): 257-75; David A. Gamson, Xiaofei Lu, and Sarah Anne Eckert, "Challenging the Research Base of the Common Core State Standards: A Historical Reanalysis of Text Complexity," Educational Researcher 42, no. 7 (2013): 381-91.
61 Rayner, "Understanding Eye Movements."
62 Stephen Reder, Digital Inclusion and Digital Literacy in the United States: A Portrait from PIAAC's Survey of Adult Skills (Washington, DC: American Institutes for Research, 2015).
63 Workforce Innovation and Opportunity Act of 2014, Pub. L. No. 113-128, 128 Stat. 1425 (2014).
64 Elementary and Secondary Education Act of 1965, Pub. L. No. 89-10, 79 Stat. 27 (1965).
65 National Research Council, Improving Adult Literacy.
66 Greenberg, Ehri, and Perin, "Word-Reading Processes"; Greenberg, Ehri, and Perin, "Adult Literacy Students."
67 See also David Braze, Whitney Tabor, Donald P. Shankweiler, and W. Einar Mencl, "Speaking up for Vocabulary: Reading Skill Differences in Young Adults," Journal of Learning Disabilities 40, no. 3 (2007): 226-43; Ryan Hall, Daphne Greenberg, Jacqueline Laures-Gore, and Hye K. Pae, "The Relationship between Expressive Vocabulary Knowledge and Reading Skills for Adult Struggling Readers," Journal of Research in Reading 37 (2014): 87-100; Hye K. Pae, Daphne Greenberg, and Robin D. Morris, "Construct Validity and Measurement Invariance of the Peabody Picture Vocabulary Test-III in the Performance of Struggling Adult Readers," Language Assessment Quarterly 9 (2012): 152-71; Hye K. Pae, Daphne Greenberg, and Rihana S. Williams, "An Analysis of Differential Error Patterns on the Peabody Picture Vocabulary Test-III in Children and Struggling African-American Adult Readers," Reading and Writing: An Interdisciplinary Journal 25 (2012): 1239-58; Nancy L. To, Elizabeth L. Tighe, and Katherine S. Binder, "Investigating Morphological Awareness and the Processing of Transparent and Opaque Words in Adults with Low Literacy Skills and in Skilled Readers," Journal of Research in Reading (2015): 1-18, advance online publication, http://onlinelibrary.wiley.com/doi/10.1111/1467-9817.12036/full.
68 For example, Joanne F. Carlisle, "Awareness of the Structure and Meaning of Morphologically Complex Words: Impact on Reading," Reading and Writing 12 (2000): 169-90; Joanne F. Carlisle and C. Addison Stone, "The Effects of Morphological Structure on Children's Reading of Derived Words," in Reading Complex Words: Cross-Language Studies, ed. Egbert M. H. Assink and Dominick Sandra (London: Kluwer Academic, 2003); Anne E. Fowler and Isabelle Y. Liberman, "The Role of Phonology and Orthography in Morphological Awareness," in Morphological Aspects of Language Processing, ed. Laurie Beth Feldman (Hillsdale, NJ: Erlbaum, 1995), 157-88; Tiffany P. Hogan, Mindy Sittner Bridges, Laura M. Justice, and Kate Cain, "Increasing Higher Level Language Skills to Improve Reading Comprehension," Focus on Exceptional Children 44 (2011): 1-20; Li-jen Kuo and Richard C. Anderson, "Morphological Awareness and Learning to Read: A Cross-Language Perspective," Educational Psychologist 41 (2006): 161-80; Xiuli Tong, S. Helene Deacon, John R. Kirby, Kate Cain, and Rauno Parrila, "Morphological Awareness: A Key to Understanding Poor Reading Comprehension in English," Journal of Educational Psychology 103 (2011): 523-34.
69 For example, Virginia Mann, Donald Shankweiler, and Suzanne T. Smith, "The Association between Comprehension of Spoken Sentences and Early Reading Ability: The Role of Phonetic Representations," Journal of Child Language 11 (1984): 627-43.
70 For example, Erica McClure and Margaret S. Steffensen, "A Study of the Use of Conjunctions across Grades and Ethnic Groups," Research in the Teaching of English 19 (1985): 217-36.
71 Nicole A. Taylor, Daphne Greenberg, Jacqueline Laures-Gore, and Justin C. Wise, "Exploring the Syntactic Skills of Struggling Adult Readers," Reading and Writing: An Interdisciplinary Journal 25 (2012): 1385-402.
72 Curtis and Kruidenier, Teaching Adults to Read.
73 Danielle. S. McNamara, ed., Reading Comprehension Strategies: Theories, Interventions, and Technologies (Mahwah, NJ: Lawrence Erlbaum, 2007).
74 Michelene T. H. Chi, Nicholas de Leeuw, Mei-Hung Chiu, and Christian LaVancher, "Eliciting Self-Explanations Improves Understanding." Cognitive Science 18, no. 3 (1994): 439-77; Danielle S. McNamara, "SERT: Self-Explanation Reading Training," Discourse Processes 38, no. 1 (2004): 1-30; Sally A. Radmacher and Elizabeth Latosi-Sawin, "Summary Writing: A Tool to Improve Student Comprehension and Writing in Psychology," Teaching of Psychology 22 (1995): 113-15.
75 Patricia F. Vadasy and Elizabeth A. Sanders, "Repeated Reading Intervention: Outcomes and Interactions with Readers' Skills and Classroom Instruction," Journal of Educational Psychology 100, no. 2 (2008): 272-90.
76 Sabatini et al., "Reading Intervention Programs"; Jane Shore, John Sabatini, Jennifer Lentini, Steven Holtzman, and Adjua McNeil, "Development of an Evidence-Based Reading Fluency Program for Adult Literacy Learners," Reading Psychology 36, no. 1 (2015): 86-104.
77 Alamprese et al., "Structured Decoding Curriculum"; Jennifer G. Cromley, "Metacognition, Cognitive Strategy Instruction, and Reading in Adult Literacy," in Review of Adult Learning and Literacy: Connecting Research, Policy, and Practice, ed. John Comings, Barbara Garner and Christine Smith (Mahwah, NJ: Lawrence Erlbaum, 2005), 187-204; Curtis and Kruidenier, Teaching Adults to Read; Greenberg, "Struggling Adult Readers"; Hock and Mellard, "Learning Strategies Instruction"; Victoria Purcell-Gates, Erik Jacobson, and Sophie Degener, Print Literacy Development: Uniting the Cognitive and Social Practice Theories (Cambridge, MA: Harvard University Press, 2004); Stephen Reder, The Longitudinal Study of Adult Learning: Challenging Assumptions (Montreal: Centre for Literacy, 2012), http://www.centreforliteracy.qc.ca/sites/default/files/CFLRsrchBrief_Chllngng_Assmptns.pdf; Sabatini et al., "Reading Intervention Programs"; also see National Research Council, Improving Adult Literacy.
78 Juliana M Taymans, ed., Learning to Achieve: A Review of the Research Literature on Serving Adults with Disabilities (Washington DC: National Institute for Literacy, 2009).