View in EDS HTML Full Text PDF Full Text

A Closer Look at Low-Performing Adult Readers in the Scandinavian Countries

Saved in:
Bibliographic Details
Title: A Closer Look at Low-Performing Adult Readers in the Scandinavian Countries
Language: English
Authors: Gabrielsen, Egil (ORCID 0000-0003-4296-5591), Sabatini, John (ORCID 0000-0002-0292-2039)
Source: Scandinavian Journal of Educational Research. 2020 64(2):242-255.
Availability: Routledge. Available from: Taylor & Francis, Ltd. 530 Walnut Street Suite 850, Philadelphia, PA 19106. Tel: 800-354-1420; Tel: 215-625-8900; Fax: 215-207-0050; Web site: http://www.tandf.co.uk/journals
Peer Reviewed: Y
Page Count: 14
Publication Date: 2020
Document Type: Journal Articles
Reports - Research
Descriptors: Adult Literacy, Low Achievement, Adults, Reading Skills, Reading Tests, Foreign Countries, Sentences, Reading Comprehension, Decoding (Reading), Printed Materials, Time on Task, Basic Vocabulary, Scores
Geographic Terms: Denmark, Norway, Sweden
Assessment and Survey Identifiers: Program for the International Assessment of Adult Competencies (PIAAC)
DOI: 10.1080/00313831.2018.1539033
ISSN: 0031-3831
Abstract: This article focuses on low-performing adult readers in the Scandinavian countries, seeking to identify and describe some of the literacy-related problems that they face. Following a presentation of the Reading Components Assessment (RCA) used in PIAAC (Program for International Assessment of Adult Competencies), we provide profiles for those informants who completed the RCA. Then we compare the RCA results of adults having well-functioning literacy skills with the results of adults identified as low-performing readers. The presentation includes both results for items correct and information about the time used to complete the items. The findings indicate that the gap in performance on the Component task sets is wider between adults scoring "Below level 1" in literacy and those scoring on Level 1 than the corresponding gap between "Level 1" and "Level 2." Implications for policy and practice are discussed.
Abstractor: As Provided
Entry Date: 2020
Accession Number: EJ1240868
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwHo0h_Fu6TyyTkqH_L6grUqAAAA4jCB3wYJKoZIhvcNAQcGoIHRMIHOAgEAMIHIBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDLWsulvFn8Q1sHiILAIBEICBmj-6pia9xb-WYx-Z7KF4cLcZW4ZXqVgVYi2PfYa_ZeQKjsmL448X3o9_U9wj-zdA2BNSGKP1EzxUxmbE0v1Wu0VV60eTtOfWmCUNt6R15GmV4A3miKeO9iXRxEqnc0qieNlnvW3wjcnacdQm1HdxP7lbWOg8BhgVsFcLA0pTLRi-9rTJh_kbI1j-R1gAGTAp3iCGu8yzpBD7S6o=
Text:
  Availability: 1
  Value: <anid>AN0141337647;55b01mar.20;2020Jan24.02:28;v2.2.500</anid> <title id="AN0141337647-1">A Closer Look at Low-Performing Adult Readers in the Scandinavian Countries </title> <p>This article focuses on low-performing adult readers in the Scandinavian countries, seeking to identify and describe some of the literacy-related problems that they face. Following a presentation of the Reading Components Assessment (RCA) used in PIAAC (Program for International Assessment of Adult Competencies), we provide profiles for those informants who completed the RCA. Then we compare the RCA results of adults having well-functioning literacy skills with the results of adults identified as low-performing readers. The presentation includes both results for items correct and information about the time used to complete the items. The findings indicate that the gap in performance on the Component task sets is wider between adults scoring Below level 1 in literacy and those scoring on Level 1 than the corresponding gap between Level 1 and Level 2. Implications for policy and practice are discussed.</p> <p>Keywords: Low-performing adult readers; PIAAC; reading components; print vocabulary; sentence processing; passage comprehension</p> <hd id="AN0141337647-2">Introduction</hd> <p>Adults with a limited literacy ability are at a distinct disadvantage in today's society. The ability to read and learn from the text is a foundation skill for adult citizens, essential for mastering their life: for finding and retaining employment, for functioning in daily life and for maintaining good health and well-being. Adults with weak reading skills often opt out of the literacy environment; they may be slow and inaccurate readers and tend to avoid reading activities, which creates a vicious cycle for their reading development.</p> <p>Program for International Assessment of Adult Competencies (PIAAC) assesses the proficiency of adults in literacy, numeracy and problem solving in technology-rich environments. These skills are seen as "key information-processing competencies" that are relevant to adults in many social contexts and work situations, and necessary for fully integrating and participating in the labor market, education and training, and social and civic life. Around 1,66,000 adults aged 16–65 were surveyed in 24 countries during the first cycle of PIAAC in 2012. 16,700 were from the Scandinavian countries (OECD, [<reflink idref="bib12" id="ref1">12</reflink>]).</p> <p>The Scandinavian countries (Denmark, Norway and Sweden) have many social, political and cultural common features. One of them is free education and they are among the countries with the highest share in higher education. They have been among the best overall performers in the international surveys measuring adult competencies. Even so, many low-performing adult readers were identified in Scandinavia, both in IALS[<reflink idref="bib1" id="ref2">1</reflink>] (OECD & Statistics Canada, [<reflink idref="bib10" id="ref3">10</reflink>]) and – in even larger numbers – in PIAAC. This article will focus on the low-performing adult readers in the Scandinavian countries and try to identify and describe some of the literacy-related problems which they face.</p> <p>Compared with IALS, PIAAC gathered more detailed information about adults performing at the lower end of the literacy scale. IALS defined <emph>Level 1</emph> as the lowest level of literacy performance (out of five levels), but PIAAC introduced an additional level: <emph>Below level 1</emph>. (See appendix for more details about Level 1 and Below level 1). Further, PIAAC informants who took the paper-based version of the survey were given an additional test of basic reading components (OECD, [<reflink idref="bib12" id="ref4">12</reflink>]).</p> <p>Before introducing results, we will present an overview of the Reading Components Assessment used in PIAAC: the principles for constructing the test, a brief description of scales and items included in the test, and finally an overview of the informants who took this test.</p> <p>Against this background, we will then take a closer look at the low-performing readers to answer a few questions about them: How many are they in the Scandinavian countries, compared to the numbers in other countries participating in PIAAC? What challenges does reading involve for them? How do they perform on the various Reading Component task sets compared with adults who have been defined as skilled readers? From this can be extracted the two main themes of this article:</p> <p></p> <ulist> <item> <emph>Validity</emph>: To what extent does the Reading Components Assessment in PIAAC give us a broader picture of low-performing adult readers in the Scandinavian countries?</item> <p></p> <item> <emph>Comparison of results on the Component task sets</emph>: How do low-performing and high-performing readers in Scandinavia compare with each other?</item> </ulist> <hd id="AN0141337647-3">The Reading Components Assessment in PIAAC</hd> <p>A higher level of literacy skills builds on a foundation of fluent component skills (Perfetti, [<reflink idref="bib14" id="ref5">14</reflink>]; Sabatini, [<reflink idref="bib16" id="ref6">16</reflink>]). Reading components are indicators of a person's competencies with regard to the most fundamental elements of understanding written texts: the ability to recognize and understand printed words and to build meaning from sentences and passages. It has been claimed that reading components make up the basic set of decoding skills that are essential for extracting meaning from written texts (OECD, [<reflink idref="bib12" id="ref7">12</reflink>]). This basic principle of learning to read has been widely accepted internationally (e.g., Curtis, [<reflink idref="bib3" id="ref8">3</reflink>]; Oakhill, Cain, & Bryant, [<reflink idref="bib9" id="ref9">9</reflink>]; Perfetti, [<reflink idref="bib14" id="ref10">14</reflink>]; Sabatini, [<reflink idref="bib16" id="ref11">16</reflink>]).</p> <p>More detailed information about these skills might help us identify the aspects of reading that low-literate adults can and cannot manage, thus providing us with a necessary basis for helping people with insufficient literacy skills. Literacy-proficiency scores alone are not sufficient for determining how adults can become more fluent and efficient readers. Hence reading components can be seen as a bridge to a deeper understanding of the needs and capabilities of adults who score at or below <emph>Level 1</emph> in surveys such as PIAAC.</p> <p>For most countries participating in PIAAC, 95% or more of adults between the ages of 16 and 65 were assessed as having skills that extend well beyond the ability to decode and comprehend basic texts.</p> <p>Our concern here is for the others: those adults who have difficulty even with the most basic applied-literacy items. The question that arises is whether these individuals are entirely illiterate or possess some reading skills. In the previous surveys of adult literacy, IALS 2000 and ALL[<reflink idref="bib2" id="ref12">2</reflink>] 2007, respondents who could not answer any of the literacy items correctly were assigned a score at the lowest limit of the literacy scale. In PIAAC, by contrast, the 23 participating countries were invited by the OECD to include a Reading Components Assessment to obtain more information about those respondents.</p> <p>The reading component items were scaled concurrently with the literacy items in order to place them on a common scale. All the items were fit using a multigroup extension of the two-parameter logistic model (2PL; Birnbaum, [<reflink idref="bib1" id="ref13">1</reflink>]; Bock & Zimowski, [<reflink idref="bib2" id="ref14">2</reflink>]). Due to the matrix sampling design for PIAAC, scores are not estimated for each individual; rather, plausible values are drawn randomly from estimated country-specific ability distributions for students with similar item-response patterns and background characteristics (Von Davier, Gonzalez, & Mislevy, [<reflink idref="bib21" id="ref15">21</reflink>]). While plausible values are not the best estimate of proficiency for any one individual, in the aggregate they provide the best estimate of the population distribution. For this reason, plausible values have become essential components of most data analyses conducted using large-scale assessment data in education. That said, the reliability of the scores, across countries is quite good with estimates ranging from.84 to.90 with a mean of.87. The accuracy of the reading components in PIAAC was also quite good. Results show a high proportion of correct responses (vocabulary: P+ = 97.4; sentence processing: P+ = 91.4; basic passage comprehension: P+ = 92.0). More information about the scaling and the psychometric properties of the PIAAC scales can be found in the PIAAC technical report (OECD, [<reflink idref="bib13" id="ref16">13</reflink>]).</p> <hd id="AN0141337647-4">Principles for the Construction of the Reading Components Assessment in PIAAC</hd> <p>Hoover and Tunmer ([<reflink idref="bib5" id="ref17">5</reflink>]) introduced the simple view of reading with two claims: first, that reading consists of word recognition and linguistic comprehension; and second, that while each of these is necessary for reading, neither is sufficient in itself (p. 3). Word recognition is the strongest predictor of reading level in the early years of reading development and continues to contribute significant variance in skilled readers as well. However, when word recognition becomes more automatized, listening comprehension is a stronger predictor of a person's reading ability (McCormick, [<reflink idref="bib7" id="ref18">7</reflink>]). Strucker, Yamamoto, and Kirsch ([<reflink idref="bib20" id="ref19">20</reflink>]) introduced a similar component framework, using the terms print components (e.g., decoding accuracy and fluency) and meaning components (e.g., oral vocabulary).</p> <p>It is important to underline that the components mentioned above are integrated in skilled reading and do not develop in a strictly hierarchical way during the period when we learn to read. This means that a child does not wait to "master" decoding skills until he or she has learned to construct meaning (Sabatini & Bruce, [<reflink idref="bib17" id="ref20">17</reflink>]).</p> <p>The first principle described in the Reading Components Assessment framework is that comprehension, defined as the <emph>meaning-construction processes</emph> of reading, is based on knowledge of how a language is represented in its specific writing system (Sabatini & Bruce, [<reflink idref="bib17" id="ref21">17</reflink>]). On an individual level, this can be assessed by using tasks that examine ability and efficiency in processing the elements of the written language. This includes letters/characters, words, sentences and larger text segments.</p> <p>A second principle presented in this framework is that the tasks included in the test are <emph>not</emph> designed to separately measure a person's level of language skills in the writing system. It is therefore not possible to disentangle low language skills from low literacy skills in the Reading Components Assessment (Sabatini & Bruce, [<reflink idref="bib17" id="ref22">17</reflink>], p. 6). This means that non-native speakers of the target language who do not have a basic vocabulary and basic syntactic/grammatical and linguistic comprehension skills will attain low scores.</p> <p>A third principle in this model of reading is that the level of proficiency, efficiency and integration of component skills can indicate a person's level and learning potential in reading development (Sabatini & Bruce, [<reflink idref="bib17" id="ref23">17</reflink>], p. 6). Components efficiency is measured by speed and rate of processing as well as accuracy.</p> <p>In addition to the three principles presented above, the framework also includes two guiding assumptions for the assessment of component skills. The first stresses that the choice of item types and difficulties is supposed to function for readers at the low end of the literacy spectrum. The other assumption is that, since the relationship of the language to the writing system is different in different languages, the nature of the items used to assess the components must be adapted based on consideration of those differences. Each participating country was therefore given a base set of Components items in English and guidelines for translation and adaptation for their local version of the Components test (Sabatini & Bruce, [<reflink idref="bib17" id="ref24">17</reflink>], p. 7).</p> <hd id="AN0141337647-5">Presentation of Scales and Items</hd> <p>The meaning-based skills assessed in the PIAAC Reading Components Assessment are <emph>print vocabulary</emph>, <emph>sentence processing</emph> and <emph>basic passage comprehension</emph>. This assessment was completed by 20 of the 23 countries participating in PIAAC 2012; Cyprus, Finland, and Japan did not include this part of the survey (OECD, [<reflink idref="bib12" id="ref25">12</reflink>]).</p> <p>One barrier to understanding what one reads is not knowing the meaning of the words in a given text. While it is possible to infer the meaning of unknown words from their context, this often results in uncertain and incomplete word meanings. In addition, this is a time-consuming strategy.</p> <p>The <emph>Print Vocabulary</emph> task set seeks to determine whether a respondent can identify words (in print) from the everyday listening lexicon of an average adult speaker of the language. This would, for example, be the language used in newspapers, radio, and television. The choice of items emphasizes everyday words in the language which are seen as the most cross-country, comparable vocabulary.</p> <p>The purpose of the vocabulary component measure [...] is not so much to measure the full extent of individual's vocabulary knowledge, so much as determining whether individuals could understand words when reading print that they could otherwise understand when listening to those words. (Sabatini & Bruce, [<reflink idref="bib17" id="ref26">17</reflink>], p. 10)</p> <p>The respondents were asked to connect a picture of an object to one of four alternative words. The task set consisted of 34 items, and the test takers had to circle the word that matched the picture (an example item is shown in Figure 1). Two items were presented on each page of the test booklet.</p> <p>Graph: Figure 1. Example item from the Print Vocabulary task set.</p> <p>The <emph>Sentence Processing</emph> task set requires the respondents to evaluate whether a sentence makes logical sense in terms of the properties of the real world or the internal logic of the sentence. The focus in the sentence task was entirely on meaning judgement. Each of the 22 sentences included in the test required some syntactic and semantic processing.</p> <p>Earlier studies (e.g., Kintsch, [<reflink idref="bib6" id="ref27">6</reflink>]) show that the sentence is a natural breakpoint in the reading of continuous text. A skilled reader will generally make a pause at the end of each sentence, and a variety of operations are typically performed. These include making inferences and relating meaning to background knowledge and to previous memory of the passage as it unfolds.</p> <p>This item type thus measures a combination of basic sentence meaning processing, as well as a basic comprehension monitoring and evaluation (Sabatini & Bruce, [<reflink idref="bib17" id="ref28">17</reflink>], p. 12).</p> <p>Sentences of increasing difficulty, as indexed by length, were included in this assessment. Difficulty was manipulated by adding informational density (e.g., more phrases and clauses) and by changing the conditions or relations that needed to be processed for construction of the sentence meaning. In item design, an attempt was made to minimize ambiguity of meaning, and all items were field-tested to confirm that adults could accurately make the judgment called for in the task (example items are shown in Figure 2).</p> <p>Graph: Figure 2. Example items from the Sentence Processing task set.</p> <p>The <emph>Passage Comprehension</emph> task set is based on research showing that a silent reading assessment can function as an indicator of basic reading fluency and comprehension (e.g., Wayman, Wallace, Wiley, Ticha, & Espin, [<reflink idref="bib22" id="ref29">22</reflink>]). The design uses a forced-choice cloze paradigm, which means that at certain points in the text, the respondent is given a choice of two words and has to select the word that makes the most sense in the context of the passage. The incorrect word should be obviously wrong to a reader with some basic comprehension skills; these distractors may be semantically or grammatically incorrect. The integration of decoding, word recognition, vocabulary and sentence processing is required to construct the basic meaning of a short passage.</p> <p>The test takers were required to read through four different prose texts that ranged in length from 159 to 278 words across two to three paragraphs. Most of the items in each of the four passages are of similar difficulty. This is in contrast to the Sentence Processing task set, where the sentence items become progressively more complex and difficult. The passages were constructed based on the kinds of text types that adults typically encounter: narrative, persuasive and expository (an example is shown in Figure 3).</p> <p>Graph: Figure 3. Example of text and items from the Passage Comprehension task set.</p> <p>There was no time limit, but the total time to complete each text was recorded. For adults with relatively good literacy skills, the total number of correct items may approach ceiling levels. The inclusion of the total time required to complete the task enables basic reading efficiency to be estimated.</p> <hd id="AN0141337647-6">Who Took the Reading Components Assessment in PIAAC?</hd> <p>PIAAC included two versions of the literacy assessment: a computer-based format administered on a laptop computer (CBA) and a paper-based format (PBA). The procedures and design for deciding whether an individual was administered the paper-based or the computer-based instruments were quite complex (OECD, [<reflink idref="bib12" id="ref30">12</reflink>]). The Reading Components test was completed by all respondents who took the PBA and by the respondents who failed the literacy and numeracy core assessment in the CBA.</p> <p>The information in Table 1 presents the different groups who completed the Reading Components test in PIAAC.</p> <p>Table 1. Percentages of informants who participated in the Reading Components Assessment in PIAAC.</p> <p> <ephtml> <table><thead valign="bottom"><tr><td /><td>All countries</td><td>Scandinavian countries</td></tr><tr><td>Group</td><td>Overall</td><td>PBA sub-Sample</td><td>Overall</td><td>PBA sub-Sample</td></tr></thead><tbody><tr><td>No computer experience</td><td char=".">6.2</td><td char=".">24.3</td><td char=".">1.5</td><td char=".">14.1</td></tr><tr><td>Some computer experience, but opted out of CBA</td><td char=".">5.7</td><td char=".">22.3</td><td char=".">5.4</td><td char=".">50.6</td></tr><tr><td>Some computer experience, but failed CBA core</td><td char=".">3.1</td><td char=".">12.1</td><td char=".">3.2</td><td char=".">30.0</td></tr><tr><td>Passed CBA core, but failed the literacy/numeracy core</td><td char=".">0.6</td><td char=".">0.0</td><td char=".">0.6</td><td char=".">0.0</td></tr><tr><td>Failed PBA core</td><td char=".">7.0</td><td char=".">6.9</td><td char=".">7.5</td><td char=".">7.6</td></tr><tr><td>Total taking PBA</td><td char=".">14.5</td><td char=".">100.0</td><td char=".">9.7</td><td char=".">100.0</td></tr><tr><td>Total taking reading components</td><td char=".">14.9</td><td char=".">56.5</td><td char=".">10.3</td><td char=".">74.7</td></tr></tbody></table> </ephtml> </p> <p>The values in the Overall column for All countries correspond to the percentage of individuals across all countries from the first round of PIAAC,[<reflink idref="bib3" id="ref31">3</reflink>] for the given row condition, who took the Reading Components. Similarly, the Overall column for the Scandinavian countries reports the percentages across the three countries.</p> <p>The values in the PBA Sub-Sample column for All countries correspond to the percentage of PBA individuals across all countries, for the given row condition, who took the Reading Components. Similarly, the PBA Sub-Sample column for the Scandinavian countries reports the percentages across the three countries for PBA test takers.</p> <p>For example, 6.2% of all test takers had no computer experience and took the Components test. Correspondingly, 24.3% of the test takers who took the PBA had no computer experience and took the Components test.</p> <p>Table 2 shows how the test takers from Scandinavia, who took the Reading components, were performing at the respective literacy levels in PIAAC. Table 2 indicates that in Denmark and Sweden, close to 40% of these respondents performed either at <emph>Level 1</emph> or <emph>Below level 1</emph> in literacy. In Norway, only 26% performed at the lowest literacy levels.</p> <p>Table 2. Percentages of Scandinavian informants taking the reading components test, by literacy level.</p> <p> <ephtml> <table><thead valign="bottom"><tr><td /><td><italic>Below level 1</italic></td><td><italic>Level 1</italic></td><td><italic>Level 2</italic></td><td><italic>Level 3</italic></td><td><italic>Level 4/5</italic></td><td><italic>N</italic></td></tr></thead><tbody><tr><td>Denmark</td><td char=".">10.3</td><td char=".">28.7</td><td char=".">40.5</td><td char=".">18.8</td><td char=".">1.7</td><td char=".">905</td></tr><tr><td>Norway</td><td char=".">5.5</td><td char=".">20.3</td><td char=".">40.3</td><td char=".">28.9</td><td char=".">4.9</td><td char=".">509</td></tr><tr><td>Sweden</td><td char=".">10.5</td><td char=".">27.8</td><td char=".">39.8</td><td char=".">19.5</td><td char=".">2.4</td><td char=".">329</td></tr><tr><td>International average</td><td char=".">8.1</td><td char=".">22.9</td><td char=".">39.8</td><td char=".">24.9</td><td char=".">4.3</td><td /></tr></tbody></table> </ephtml> </p> <p>Based on these numbers it is important to repeat that most of the adults who took the Components test did not perform at <emph>Below level 1</emph> in terms of proficiency. Table 1 shows that less than 15% of the full sample participating in PIAAC were routed down the PBA pathway. By contrast, only 2.4% of the entire cross-country sample of informants scored <emph>Below level 1</emph> on the literacy scale. It should be kept in mind that those scoring <emph>Below level 1</emph> make up the sub-population for which the Components test was originally designed. The 2.4% group represents only a small group of the informants who took the PBA version of PIAAC (Sabatini, [<reflink idref="bib15" id="ref32">15</reflink>]).</p> <p>This means that most of the adults who took the PBA version passed the literacy/numeracy core and hence completed both the Components test and a literacy or numeracy block. As can be seen in Table 2, this larger group demonstrated literacy skills across the ability spectrum in PIAAC. As expected, the Components tasks were very easy for most of them.</p> <hd id="AN0141337647-7">Conducting the Reading Components Assessment</hd> <p>The interviewers in PIAAC introduced the Components test booklet to the informant through a simple instruction. The three task sets were presented in this order: Print Vocabulary, Sentence Processing and Passage Comprehension. The interviewer read out the directions for each of the Component task sets. For all three task sets, the interviewer started a timer when the informant turned the page for the first item and stopped it when the final item was completed. The Passage Comprehension task set consisted of four passages, and here three separate times were recorded, with the first two passages combined into one time.</p> <p>Table 3 presents an overview of the total number of items per Component task set, the number of choices for each item in each Component task set and the approximate total score for an respondent answering by chance. The percentage correct reported for each participant is the total correct divided by the total possible score. All items left blank were treated as incorrect responses. This should be taken into account when interpreting the average scores for different proficiency levels.</p> <p>Table 3. Total items, number of choices and chance-level performance for Reading Component task sets.</p> <p> <ephtml> <table><thead valign="bottom"><tr><td>Reading component</td><td>Number of items</td><td>Number of choices per item</td><td>Chance-level total score</td><td>Chance-level percentage correct</td></tr></thead><tbody><tr><td>Print Vocabulary</td><td char=".">34</td><td char=".">4</td><td>8–9</td><td>24–26</td></tr><tr><td>Sentence Processing</td><td char=".">22</td><td char=".">2</td><td char=".">11</td><td char=".">50</td></tr><tr><td>Passage Comprehension</td><td char=".">44</td><td char=".">2</td><td char=".">22</td><td char=".">50</td></tr><tr><td>Total for all items</td><td char=".">100</td><td>2 or 4</td><td>41–42</td><td>41–42</td></tr></tbody></table> </ephtml> </p> <p>Sabatini ([<reflink idref="bib15" id="ref33">15</reflink>]).</p> <p>Another critical issue worth mentioning is that the translation of Component items across languages may result in different item-level difficulty (Sabatini, [<reflink idref="bib15" id="ref34">15</reflink>]). Some languages may be easier to learn than others because the print-to-sound correspondences are highly regular or transparent. Italian and Spanish are examples of languages with highly regular such correspondences while English is an example of the opposite. The Scandinavian languages (Danish, Norwegian and Swedish) are somewhere in between these extremes (Seymour, Aro, & Erskine, [<reflink idref="bib19" id="ref35">19</reflink>]).</p> <hd id="AN0141337647-8">Scandinavian Results on the Reading Components Assessment</hd> <p>Analyses were conducted using the IEA International Database Analyzer (3.1.1) and SPSS (version 25). The software is designed to apply iterations of analyses including all 10 plausible values (Von Davier et al., [<reflink idref="bib21" id="ref36">21</reflink>]). The approach is conceptually similar to regression analyses using multiple imputation. There were no missing values in the publically available dataset that was used.</p> <p>Tables 4–6 show the average percentages of correct scores for the three Component task sets, by proficiency level, for the three Scandinavian countries as well as the mean scores for the countries that included the Components test in round 1 of PIAAC. Results from informants at or above <emph>Level 3</emph> are not included in these tables, since they are all at 95% or higher on each task set.</p> <p>Table 4. Proportions of correct answers on the Print Vocabulary task set.</p> <p> <ephtml> <table><thead valign="bottom"><tr><td>Country</td><td><italic>Below level 1</italic></td><td><italic>Level 1</italic></td><td><italic>Level 2</italic></td></tr></thead><tbody><tr><td>Denmark</td><td char=".">0.91</td><td char=".">0.93</td><td char=".">0.96</td></tr><tr><td>Norway</td><td char=".">0.84</td><td char=".">0.91</td><td char=".">0.95</td></tr><tr><td>Sweden</td><td char=".">0.92</td><td char=".">0.95</td><td char=".">0.97</td></tr><tr><td>PIAAC average</td><td char=".">0.90</td><td char=".">0.95</td><td char=".">0.97</td></tr></tbody></table> </ephtml> </p> <p>Table 5. Proportions of correct answers on the Sentence Processing task set.</p> <p> <ephtml> <table><thead valign="bottom"><tr><td>Country</td><td><italic>Below level 1</italic></td><td><italic>Level 1</italic></td><td><italic>Level 2</italic></td></tr></thead><tbody><tr><td>Denmark</td><td char=".">0.74</td><td char=".">0.85</td><td char=".">0.92</td></tr><tr><td>Norway</td><td char=".">0.73</td><td char=".">0.84</td><td char=".">0.92</td></tr><tr><td>Sweden</td><td char=".">0.73</td><td char=".">0.82</td><td char=".">0.90</td></tr><tr><td>PIAAC average</td><td char=".">0.74</td><td char=".">0.85</td><td char=".">0.93</td></tr></tbody></table> </ephtml> </p> <p>Table 6. Proportions of correct answers on the Passage Comprehension task set.</p> <p> <ephtml> <table><thead valign="bottom"><tr><td>Country</td><td><italic>Below level 1</italic></td><td><italic>Level 1</italic></td><td><italic>Level 2</italic></td></tr></thead><tbody><tr><td>Denmark</td><td char=".">0.68</td><td char=".">0.79</td><td char=".">0.92</td></tr><tr><td>Norway</td><td char=".">0.64</td><td char=".">0.79</td><td char=".">0.90</td></tr><tr><td>Sweden</td><td char=".">0.69</td><td char=".">0.83</td><td char=".">0.94</td></tr><tr><td>PIAAC average</td><td char=".">0.63</td><td char=".">0.85</td><td char=".">0.94</td></tr></tbody></table> </ephtml> </p> <p>Print Vocabulary is the most basic part of the Components test. When looking at the results, it must be kept in mind that chance performance on this assessment represents about 25% correct. As might be expected, Print Vocabulary scores are highest in all three countries. The Norwegian score is significantly below those of both Denmark and Sweden in the lowest-scoring category, and it is also below the PIAAC average. Except as regards the <emph>Below level 1</emph> group, however, there are only small differences between average scores for the proficiency levels included in the table. For all PIAAC countries, it can be seen that Print Vocabulary scores increase as literacy skills increase. These results indicate that most adults, even those scoring <emph>Below level 1</emph> on the PIAAC literacy scale, know how to read frequently used words in their national language.</p> <p>The Sentence Processing tasks included sentences ranging from 3 to 26 words in length. For adults scoring <emph>Below level 1</emph>, these tasks seemed to be more challenging than the Print Vocabulary tasks were.</p> <p>As can be seen in Table 5, there are no score differences between the three Scandinavian countries, while there are significant differences in scores between informants performing on the different literacy levels in all countries. For adults scoring <emph>Below level 1</emph>, 74% of the answers were correct. The rate of correct performance rises to around 84% for adults on <emph>Level 1</emph>, while those performing on literacy <emph>Level 2</emph> chose the correct answer for more than 90% of the items (remember that 50% is the chance level). However, it is worth noting that even those scoring on <emph>Level 2</emph> made some errors on the longest, most difficult sentences.</p> <p>On the Passage Comprehension task set, where the chance level for giving the correct answer is also 50%, only around two out of three answers given by adults with reading performance <emph>Below level 1</emph> were correct. This reflects the relative challenge that reading multi-paragraph, continuous texts represents for adults at this proficiency level.</p> <p>For adults performing on literacy <emph>Level 1</emph>, the percentages rise to around 80%, and those scoring on <emph>Level 2</emph> were at or above 90% correct. Norway had the lowest-performing informants on all levels, but the differences compared with Denmark and Sweden were quite small.</p> <p>The results presented here may be seen to validate the Components test. Adults at or above <emph>Level 3</emph> (who are not included in the tables) are at or near ceiling-level performance on all three task sets. Most adults <emph>Below level 1</emph> have some print-vocabulary knowledge, but they are somewhat challenged at the sentence level and even more strongly so when reading passages for basic meaning. For adults at <emph>Level 1</emph>, passage comprehension is slightly lower than sentence performance in Denmark and Norway, but not in Sweden. <emph>Level 2</emph> performers score above 90% correct on all three task sets in all three countries.</p> <hd id="AN0141337647-9">Time to Complete Items</hd> <p>The respondents were allowed unlimited time to complete each of the Component task sets. For cross-country comparability, the total time required to complete the task sets was recorded. This is because average reading rates may vary from country to country, primarily as a function of language, writing system, and cultural variables.</p> <p>Knowing the total time required to complete the tasks makes it possible to estimate a respondent's basic reading efficiency. High-performing adult readers would be able to choose correct responses quickly, without much effort, and go on reading at a normal rate. Low- or moderate-skilled readers may possess some basic comprehension skills and score high on the Components test, but they would be less efficient at processing continuous text and hence slower than skilled readers. To produce a benchmark for the relative scaling efficiency/fluency of the low- or moderate-skilled readers across countries, the performances of a sub-sample of skilled readers in each participating country were collected.</p> <p>Tables 7–9 show average times (and the related standard deviations) for the Scandinavian countries on the three Component task sets as well as the average time + SDs for all PIAAC countries. Because the standard deviations associated with some averages are high, differences in averages should be interpreted with caution.</p> <p>Table 7. Average times on the Print Vocabulary task set, in minutes (standard deviation).</p> <p> <ephtml> <table><thead valign="bottom"><tr><td /><td><italic>Below level 1</italic></td><td><italic>Level 1</italic></td><td><italic>Level 2</italic></td><td><italic>Level 3</italic></td></tr></thead><tbody><tr><td>Denmark</td><td char="."><bold>3</bold><bold>.</bold><bold>75</bold> (2.67)</td><td char="."><bold>2</bold><bold>.</bold><bold>86</bold> (2.29)</td><td char="."><bold>2</bold><bold>.</bold><bold>23</bold> (1.68)</td><td char="."><bold>1</bold><bold>.</bold><bold>89</bold> (1.31)</td></tr><tr><td>Norway</td><td char="."><bold>3</bold><bold>.</bold><bold>67</bold> (2.27)</td><td char="."><bold>3</bold><bold>.</bold><bold>18</bold> (2.12)</td><td char="."><bold>2</bold><bold>.</bold><bold>28</bold> (1.59)</td><td char="."><bold>2</bold><bold>.</bold><bold>01</bold> (1.35)</td></tr><tr><td>Sweden</td><td char="."><bold>5</bold><bold>.</bold><bold>46</bold> (5.93)</td><td char="."><bold>2</bold><bold>.</bold><bold>88</bold> (2.28)</td><td char="."><bold>2</bold><bold>.</bold><bold>31</bold> (1.47)</td><td char="."><bold>2</bold><bold>.</bold><bold>06</bold> (1.27)</td></tr><tr><td>PIAAC average</td><td char="."><bold>4</bold><bold>.</bold><bold>06</bold> (2.99)</td><td char="."><bold>2</bold><bold>.</bold><bold>97</bold> (2.14)</td><td char="."><bold>2</bold><bold>.</bold><bold>34</bold> (1.54)</td><td char="."><bold>2</bold><bold>.</bold><bold>01</bold> (1.11)</td></tr></tbody></table> </ephtml> </p> <p>Table 8. Average times on the Sentence Processing task set, in minutes (standard deviation).</p> <p> <ephtml> <table><thead valign="bottom"><tr><td /><td><italic>Below level 1</italic></td><td><italic>Level 1</italic></td><td><italic>Level 2</italic></td><td><italic>Level 3</italic></td></tr></thead><tbody><tr><td>Denmark</td><td char="."><bold>3</bold><bold>.</bold><bold>93</bold> (2.80)</td><td char="."><bold>3</bold><bold>.</bold><bold>12</bold> (2.24)</td><td char="."><bold>2</bold><bold>.</bold><bold>57</bold> (1.65)</td><td char="."><bold>2</bold><bold>.</bold><bold>40</bold> (1.28)</td></tr><tr><td>Norway</td><td char="."><bold>4</bold><bold>.</bold><bold>84</bold> (3.08)</td><td char="."><bold>3</bold><bold>.</bold><bold>92</bold> (2.50)</td><td char="."><bold>2</bold><bold>.</bold><bold>87</bold> (1.61)</td><td char="."><bold>2</bold><bold>.</bold><bold>07</bold> (1.39)</td></tr><tr><td>Sweden</td><td char="."><bold>4</bold><bold>.</bold><bold>97</bold> (4.02)</td><td char="."><bold>3</bold><bold>.</bold><bold>58</bold> (2.69)</td><td char="."><bold>2</bold><bold>.</bold><bold>91</bold> (1.78)</td><td char="."><bold>2</bold><bold>.</bold><bold>57</bold> (1.29)</td></tr><tr><td>PIAAC average</td><td char=".">4.30 (2.71)</td><td char=".">3.45 (2.07)</td><td char=".">2.79 (1.50)</td><td char=".">2.40 (1.18)</td></tr></tbody></table> </ephtml> </p> <p>Table 9. Average times on the Passage Comprehension task set, in minutes (standard deviation).</p> <p> <ephtml> <table><thead valign="bottom"><tr><td /><td><italic>Below level 1</italic></td><td><italic>Level 1</italic></td><td><italic>Level 2</italic></td><td><italic>Level 3</italic></td></tr></thead><tbody><tr><td>Denmark</td><td char="."><bold>8</bold><bold>.</bold><bold>21</bold> (5.50)</td><td char="."><bold>6</bold><bold>.</bold><bold>64</bold> (3.91)</td><td char="."><bold>5</bold><bold>.</bold><bold>55</bold> (2.75)</td><td char="."><bold>4</bold><bold>.</bold><bold>74</bold> (2.09)</td></tr><tr><td>Norway</td><td char="."><bold>10</bold><bold>.</bold><bold>52</bold> (5.70)</td><td char="."><bold>7</bold><bold>.</bold><bold>92</bold> (4.99)</td><td char="."><bold>6</bold><bold>.</bold><bold>06</bold> (3.27)</td><td char="."><bold>5</bold><bold>.</bold><bold>35</bold> (2.48)</td></tr><tr><td>Sweden</td><td char="."><bold>9</bold><bold>.</bold><bold>96</bold> (6.86)</td><td char="."><bold>7</bold><bold>.</bold><bold>90</bold> (4.97)</td><td char="."><bold>6</bold><bold>.</bold><bold>19</bold> (3.07)</td><td char="."><bold>5</bold><bold>.</bold><bold>18</bold> (2.46)</td></tr><tr><td>PIAAC average</td><td char="."><bold>8</bold><bold>.</bold><bold>92</bold> (6.02)</td><td char="."><bold>7</bold><bold>.</bold><bold>49</bold> (4.42)</td><td char="."><bold>6</bold><bold>.</bold><bold>09</bold> (2.97)</td><td char="."><bold>5</bold><bold>.</bold><bold>15</bold> (2.31)</td></tr></tbody></table> </ephtml> </p> <p>For the lowest-performing group (<emph>Below level 1</emph>), it can be seen from Table 7 that Danes and Norwegians devoted less time to reading than the PIAAC average while Swedes were far above that average. For the <emph>Level 1</emph> group, Denmark and Sweden are slightly below the PIAAC average while Norway is above it. For <emph>Levels 2 and 3</emph>, the time differences across country groups are very small.</p> <p>Table 8 shows that the Danish respondents used less time at the three lowest literacy levels included in the table. The lowest-performing Norwegians and Swedes used more time than the PIAAC average. For <emph>Levels 2 and 3</emph>, even though there are some time differences across country groups, the samples are so small that their significance cannot be tested.</p> <p>Table 9 displays the same tendency as seen above: Danes used less time at all literacy levels. Norwegians and Swedes also devoted less time to reading than the PIAAC average, with one exception in the case of Norway. This probably reflects the fact that Scandinavians have languages with an easier sight-to-sound connection than, for example, the many English-speaking participants in PIAAC.</p> <p>Table 10 shows the average times for each of the levels as ratios using <emph>Level 3</emph> as baseline. The pattern is consistent with the theory: as the proficiency level increases, the time to complete the task set decreases. In other words, proficiency in reading components is related to efficiency of processing. It can be noted that adults <emph>Below level 1</emph> took between 1.73 and 2.65 times longer to complete each Component task set than <emph>Level 3</emph> readers. Adults at <emph>Level 1</emph> took 1.40 to 1.58 times longer than adults at <emph>Level 3</emph>.</p> <p>Table 10. Reading Component timer means in minutes as ratios with <emph>Level 3</emph> as baseline.</p> <p> <ephtml> <table><thead valign="bottom"><tr><td>Component</td><td><italic>Below level 1</italic></td><td><italic>Level 1</italic></td><td><italic>Level 2</italic></td><td><italic>Level 3</italic></td></tr></thead><tbody><tr><td>Print Vocabulary</td><td>Denmark 1.98 Norway 1.82 Sweden 2.65 PIAAC average 1.97</td><td>Denmark 1.51 Norway 1.58 Sweden 1.40 PIAAC average 1.45</td><td>Denmark 1.18 Norway 1.18 Sweden 1.12 PIAAC average 1.15</td><td>Denmark 1.00 Norway 1.00 Sweden 1.00 PIAAC average 1.00</td></tr><tr><td>Sentence Processing</td><td>Denmark 1.90 Norway 1.88 Sweden 1.94 PIAAC average 1.75</td><td>Denmark 1.51 Norway 1.52 Sweden 1.40 PIAAC average 1.43</td><td>Denmark 1.24 Norway 1.12 Sweden 1.13 PIAAC average 1.17</td><td>Denmark 1.00 Norway 1.00 Sweden 1.00 PIAAC average 1.00</td></tr><tr><td>Passage Comprehension</td><td>Denmark 1.73 Norway 1.97 Sweden 1.92 PIAAC average 1.73</td><td>Denmark 1.40 Norway 1.48 Sweden 1.52 PIAAC average 1.46</td><td>Denmark 1.17 Norway 1.13 Sweden 1.20 PIAAC average 1.18</td><td>Denmark 1.00 Norway 1.00 Sweden 1.00 PIAAC average 1.00</td></tr></tbody></table> </ephtml> </p> <hd id="AN0141337647-10">Associations between Reading Components and Level of Literacy Proficiency</hd> <p>An alternative approach to examining the relationship between reading components and the literacy-proficiency level is to look at how scores on the Component tasks relate to literacy scores in PIAAC, both in isolation and in combination. The regression models used in Table 11 have been computed using the entire ability distribution of those who took the paper version of the test, not only adults <emph>Below level 1</emph>. Plausible literacy-value scores were included as dependent variables in the analyses. This Table 11 shows the variance in plausible literacy-value scores accounted for by Print Vocabulary (PV), Sentence Processing (SP), and Passage Comprehension (PC), separately and in combination, and finally with timer information included.</p> <p>Table 11. Univariate and multiple regression model adjusted <emph>r</emph>-squared values with plausible literacy-value scores as dependent variable and combinations of reading components (RCs) as independent variables</p> <p> <ephtml> <table><thead valign="bottom"><tr><td>Country</td><td>Print Vocabulary (PV)</td><td>Sentence Processing (SP)</td><td>Passage Comprehension (PC)</td><td>PV+SP</td><td>PV+SP+PC</td><td>ALL RCs + Timer</td></tr></thead><tbody><tr><td>Denmark</td><td char=".">0.01</td><td char=".">0.08</td><td char=".">0.08</td><td char=".">0.09</td><td char=".">0.11</td><td char=".">0.27</td></tr><tr><td>Norway</td><td char=".">0.07</td><td char=".">0.12</td><td char=".">0.13</td><td char=".">0.12</td><td char=".">0.15</td><td char=".">0.27</td></tr><tr><td>Sweden</td><td char=".">0.10</td><td char=".">0.17</td><td char=".">0.16</td><td char=".">0.17</td><td char=".">0.21</td><td char=".">0.35</td></tr><tr><td>PIAAC average</td><td char=".">0.04</td><td char=".">0.14</td><td char=".">0.13</td><td char=".">0.14</td><td char=".">0.17</td><td char=".">0.30</td></tr></tbody></table> </ephtml> </p> <p>The logic of order entry of variables in the regression models is based on theoretical models of reading processing (Kintsch, [<reflink idref="bib6" id="ref37">6</reflink>]; Perfetti, [<reflink idref="bib14" id="ref38">14</reflink>]). Specifically, the idea is that, as one reads, words are recognized and their meanings are accessed (enter PV first). Then the meanings of sentence propositions are encoded (enter SP second). As sentence propositions are encoded, they are integrated with the overall passage meaning or gist (enter PC third). Finally, as proficiency in reading increases, these cognitive processes become increasingly automatized: they occur with accuracy, speed, and ease, such that minimal attentional resources are consumed (Perfetti, [<reflink idref="bib14" id="ref39">14</reflink>]). Thus, this sequence best captures the processing levels of reading. As already observed, the average correct score converges at around <emph>Level 3</emph>. Hence it can be inferred that most of the predictive value of reading components will concern the description of scores below that level.</p> <p>It is clear from Table 11 that Print Vocabulary has the weakest relationship to literacy scores in all Scandinavian countries. This is likely because of the near-ceiling performance levels seen beyond <emph>Level 2</emph>. Sentence Processing and Passage Comprehension both show a stronger relationship with literacy scores – relatively strongest in Sweden and weakest in Denmark. However, combining all three reading components increases the strength of the relationship in all three countries.</p> <p>Further, including information about the time taken to complete the task sets makes a significant contribution to the relationship, with an international average of 30% of the variance accounted for by the Component scores. The corresponding percentage is a little lower than the PIAAC average in Denmark and Norway (27%) but higher in Sweden (35%).</p> <p>The relationships presented here are consistent with the theoretical literature and confirm that reading components are foundational skills necessary to attain higher-proficiency skills in reading. Further, these relationships indicate that as literacy proficiency improves, the fluency and automaticity of the underlying foundation skills increase as well; this holds true across the entire ability distribution.</p> <hd id="AN0141337647-11">Discussion</hd> <p>All countries participating in PIAAC have a sizeable adult population with literacy needs. The inclusion of the Reading Components Assessment revealed that there is great variation in the characteristics of low-performing adult readers. The proportion of adults scoring at or below <emph>Level 1</emph> in literacy skills differs a great deal across countries. Japan has the smallest share (4.9%) of adults at these levels in literacy whereas Italy (27.7%) and Spain (27.5%) have the largest. In the Nordic countries, the proportion of low-performing adults ranges from 10.7% in Finland to 15.7% in Denmark, with Norway at 12.3% and Sweden at 13.3% in between (OECD, [<reflink idref="bib12" id="ref40">12</reflink>]).</p> <p>The results of the three Scandinavian countries (Denmark, Norway, and Sweden) on the Reading Components Assessment are generally well in line with the average of the 20 countries that participated in this part of the survey. The average performance on the Print Vocabulary task set of adults scoring <emph>Below level 1</emph> in literacy was well above chance performance in all three countries (between 84% and 92%). For adults scoring at <emph>Level 1</emph>, the corresponding percentages were higher (between 91% and 95%), and for informants at <emph>Level 2</emph> they were 95% or above. These results confirm that even adults at the lowest literacy level are able to recognize printed forms of words associated with common objects such as shapes, items of furniture and animals.</p> <p>Many adults scoring <emph>Below level 1</emph> in literacy are also able to understand and evaluate short sentences, at least when the complexity of the text-and-task combination is constrained. The percentage of correct answers on the Sentence Processing task set for this group was 73% to 74% in the Scandinavian countries – well in excess of chance performance, which would be about 50%. The corresponding scores were 82% or above for informants at <emph>Level 1</emph> and 90% or above for those at <emph>Level 2</emph>.</p> <p>However, adults performing <emph>Below level 1</emph> in literacy have greater difficulties when it comes to processing complex sentences and reading extended passages for basic meaning. The Scandinavian results on the Passage Comprehension task set for this group of respondents were much closer to chance levels (i.e., 50%): between 64% and 69%, as against around 80% for adults performing on <emph>Level 1</emph> in literacy and 90% or above for those on <emph>Level 2</emph>.</p> <p>In line with theories about the automaticity of processes involved in reading, we found an association between the time required to complete items and accuracy results. Specifically, as the total percentage of correct answers increased, the time required to complete the items decreased. That is, there is an association between proficiency and efficiency (Perfetti, [<reflink idref="bib14" id="ref41">14</reflink>]). Proficient readers' speed, accuracy and ease of processing is interpreted to reflect that they are able to conserve attention resources, which can then be used to meet higher-order processing demands, such as those involved in solving complex literacy tasks.</p> <p>In conclusion, scores on the Reading Components Assessment provide a more detailed profile of the foundational reading skills which adults possess. The results presented in this article indicate that the vast majority of low-performing readers do possess some literacy skills – that they are able to engage in a limited way with texts. In other words, assumptions of illiteracy pure and simple based on survey results have been proven to be overstated. At the same time, however, it must be kept in mind that there are adults in every country who lack even the most basic foundational literacy skills, meaning that their participation in a literate world is severely limited (Grotlüschen, Mallows, Reder, & Sabatini, [<reflink idref="bib4" id="ref42">4</reflink>]; OECD, [<reflink idref="bib12" id="ref43">12</reflink>]; Sabatini, [<reflink idref="bib15" id="ref44">15</reflink>]).</p> <p>The present findings also indicate that the gap in performance on the Component task sets is wider between adults scoring <emph>Below level 1</emph> in literacy and those scoring on <emph>Level 1</emph> than the corresponding gap between <emph>Level 1</emph> and <emph>Level 2</emph>. This suggests that a primary target for policy interventions in Denmark, Norway and Sweden should be the group of around 450,000 adults in those countries who are <emph>Below level 1</emph> in literacy in PIAAC. Still, there are also many adults on <emph>Level 1</emph> who need to improve their literacy skills. This group totals more than one million people in Scandinavia.</p> <p>However, these are not homogeneous groups offering policy-makers simple and clear targets to aim for in any of the three Scandinavian countries. This should be taken into account in the planning and design of training programs for these target groups. A better understanding of the relationship between reading components and general literacy proficiency can be helpful in creating instructional programs for adults seeking to enhance their literacy abilities.</p> <p>The results from the Component task sets, in that they target foundational skills, can provide valuable new information, for policy and practice alike, beyond general information about reading-proficiency levels. Combined with what can be learned from the research literature on the teaching of adults, these results enable some conclusions to be drawn about the appropriate design of instructional programs. For example, for adults who still find it challenging to decode or recognize common words, it may be a good idea to adapt methods used with beginning readers (National Research Council, [<reflink idref="bib8" id="ref45">8</reflink>]). Another recommendation is for reading-literacy programs to include frequent opportunities for reading practice. Repeated reading of familiar texts is a technique that has been shown to help build reading fluency (Sabatini, Shore, Holtzman, & Scarborough, [<reflink idref="bib18" id="ref46">18</reflink>]).</p> <p>By contrast, solely teaching sentences in isolation is not recommended for any instructional program. The action undertaken to help low-skilled readers increase their proficiency needs to be highly contextualized so that it takes advantage not only of the broader knowledge of the world that adult learners possess compared with children, but also of each individual's relevant interests. Hence isolating all instruction to individual components is not likely to be as effective as finding the appropriate mix for each adult person.</p> <p>A good instructional program will vary the complexity of the texts that the adult reads, and it will provide opportunities to discuss short texts. It is likely that brief, continuous texts of the sort that are appropriate for discussion will cover the range of meaning relationships that are represented in the Sentence Processing items: referential, causal and knowledge-based relationships. Adult learners can discuss their understanding of texts or try to summarize or explain them to others, either to a teacher or to other adult learners.</p> <p>Adults with either inaccurate or slow passage-reading skills would benefit from more practice involving the reading of increasingly longer, continuous prose documents, both narrative and expository. The strong association between accuracy and processing speed presented in this article implies that training programs should encourage extended practice and engagement with texts to enhance the ease, speed and efficiency of processing written text.</p> <p>Finally, adults and their educators need to plan for a long-term investment of time and effort in order to reach higher levels of achievement. It is important to engage with the inter-connected, digital literacy world of our age, which besides being a key literacy arena can also offer adults wishing to improve their skills a variety of services and support tailored to their needs.</p> <hd id="AN0141337647-12">Disclosure statement</hd> <p>No potential conflict of interest was reported by the authors.</p> <hd id="AN0141337647-13">Appendix</hd> <p></p> <p> <ephtml> <table><tbody><tr><td><bold>Types of tasks completed successfully at each level of proficiency</bold></td></tr><tr><td><bold>Below level 1</bold></td></tr><tr><td>The tasks at this level require the respondent to read brief texts on familiar topics to locate a single piece of specific information. There is seldom any competing information in the text and the requested information is identical in form to information in the question or directive. The respondent may be required to locate information in short continuous texts. However, in this case, the information can be located as if the text were non-continuous in format. Only basic vocabulary knowledge is required, and the reader is not required to understand the structure of sentences or paragraphs or make use of other text features. Tasks below Level 1 do not make use of any features specific to digital texts.</td></tr><tr><td><bold>Level 1</bold> Most of the tasks at this level require the respondent to read relatively short digital or print continuous, non-continuous, or mixed texts to locate a single piece of information that is identical to or synonymous with the information given in the question or directive. Some tasks, such as those involving non-continuous texts, may require the respondent to enter personal information onto a document. Little, if any, competing information is present. Some tasks may require simple cycling through more than one piece of information. Knowledge and skill in recognizing basic vocabulary determining the meaning of sentences, and reading paragraphs of text is expected.</td></tr></tbody></table> </ephtml> </p> <p>OECD (2013b, p. 64).</p> <ref id="AN0141337647-14"> <title> References </title> <blist> <bibl id="bib1" idref="ref2" type="bt">1</bibl> <bibtext> Birnbaum, A. (1968). Some latent trait models and their use in inferring an examinee's ability. In F. M. Lord, & M. R. Novick (Eds.), Statistical theories of mental test scores (pp. 397 – 479). Reading, MA : Addison-Wesley.</bibtext> </blist> <blist> <bibl id="bib2" idref="ref12" type="bt">2</bibl> <bibtext> Bock, R. D., & Zimowski, M. F. (1997). Multiple group IRT. In W. J. van der Linden, & R. K. Hambleton (Eds.), Handbook of modern item response theory (pp. 433 – 448). New York : Springer-Verlag.</bibtext> </blist> <blist> <bibl id="bib3" idref="ref8" type="bt">3</bibl> <bibtext> Curtis, M. E. (1980). Development of components of reading skill. Journal of Educational Psychology, 72, 656 – 669. doi: 10.1037/0022-0663.72.5.656</bibtext> </blist> <blist> <bibl id="bib4" idref="ref42" type="bt">4</bibl> <bibtext> Grotlüschen, A., Mallows, D., Reder, S., & Sabatini, J. (2016). Adults with Low proficiency in literacy or numeracy (OECD Education Working Papers No. 131). Paris : OECD Publishing.</bibtext> </blist> <blist> <bibl id="bib5" idref="ref17" type="bt">5</bibl> <bibtext> Hoover, W. A., & Tunmer, W. E. (1993). The components of reading. In G. G. Thompson, W. E. Tunmer, & T. Nicholson (Eds.), Reading acquisition processes (pp. 1 – 19). Philadelphia : Multilingual Matters.</bibtext> </blist> <blist> <bibl id="bib6" idref="ref27" type="bt">6</bibl> <bibtext> Kintsch, W. (1998). Comprehension: A paradigm for cognition. Cambridge : Cambridge University Press.</bibtext> </blist> <blist> <bibl id="bib7" idref="ref18" type="bt">7</bibl> <bibtext> McCormick, S. (1994). A nonreader becomes a reader: A case study of literacy acquisition by a severely disabled reader. Reading Research Quarterly, 29, 156 – 177. doi: 10.2307/747809</bibtext> </blist> <blist> <bibl id="bib8" idref="ref45" type="bt">8</bibl> <bibtext> National Research Council. (2012). Improving adult literacy instruction: Options for practice and research. Committee on Learning Sciences: Foundations and Applications to Adolescent and Adult Literacy (A. M. Lesgold & M. Welch-Ross, Eds.). Division of Behavioral and Social Sciences and Education. Washington, DC : The National Academies Press.</bibtext> </blist> <blist> <bibl id="bib9" idref="ref9" type="bt">9</bibl> <bibtext> Oakhill, J., Cain, K., & Bryant, P. (2003). The dissociation of word reading and text comprehension: Evidence from component skills. Language and Cognitive Processes, 18 (4), 443 – 468. doi: 10.1080/01690960344000008</bibtext> </blist> <blist> <bibtext> OECD & Statistics Canada. (2000). Literacy in the information Age: Final report of the international adult literacy survey. Paris : OECD Publishing.</bibtext> </blist> <blist> <bibtext> OECD & Statistics Canada. (2011). Literacy for life: Further results from the adult literacy and life skills survey. Paris : OECD Publishing.</bibtext> </blist> <blist> <bibtext> OECD. (2013a). OECD skills outlook 2013: First results from the survey of adult skills. Paris : OECD Publishing.</bibtext> </blist> <blist> <bibtext> OECD. (2013b). Technical report of the survey of adult skills (PIAAC). Paris : OECD. Retrieved from <ulink href="http://www.oecd.org/site/piaac/surveyofadultskills.htm">http://www.oecd.org/site/piaac/surveyofadultskills.htm</ulink></bibtext> </blist> <blist> <bibtext> Perfetti, C. (2003). The universal grammar of reading. Scientific Studies of Reading, 7 (1), 3 – 24. doi: 10.1207/S1532799XSSR0701_02</bibtext> </blist> <blist> <bibtext> Sabatini, J. (2015). Understanding the basic reading skills of US adults: Policy and practical implications from the PIAAC literacy survey. ETS Center for Research on Human Capital and Education. Princeton, NJ : Educational Testing Service.</bibtext> </blist> <blist> <bibtext> Sabatini, J. P. (2002). Efficiency in word reading of adults: Ability group comparisons. Scientific Studies of Reading, 6 (3), 267 – 298. doi: 10.1207/S1532799XSSR0603_4</bibtext> </blist> <blist> <bibtext> Sabatini, J. P., & Bruce, K. M. (2009). PIAAC reading component: A conceptual framework (OECD Education Working Papers No. 33). OECD Publishing.</bibtext> </blist> <blist> <bibtext> Sabatini, J. P., Shore, J., Holtzman, S., & Scarborough, H. S. (2011). Relative effectiveness of reading intervention programs for adults with low literacy. Journal of Research on Educational Effectiveness, 4 (2), 118 – 133. doi: 10.1080/19345747.2011.555290</bibtext> </blist> <blist> <bibtext> Seymour, P. H. K., Aro, M., & Erskine, J. M. (2003). Foundation literacy acquisition in european orthographies. British Journal of Psychology, 94, 143 – 174. doi: 10.1348/000712603321661859</bibtext> </blist> <blist> <bibtext> Strucker, J., Yamamoto, K., & Kirsch, I. (2003). The relationship of the component skills of reading to IALS performance: Tipping points and five classes of adult literacy learners (NCSALL Reports No. 29). Cambridge, MA : National Center for the Study of Adult Learning and Literacy.</bibtext> </blist> <blist> <bibtext> Von Davier, M., Gonzalez, E., & Mislevy, R. (2009). What are plausible values and why are they useful. IERI Monograph Series, 2, 9 – 36.</bibtext> </blist> <blist> <bibtext> Wayman, M. M., Wallace, T., Wiley, H. I., Ticha, R., & Espin, C. A. (2007). Literature synthesis on curriculum-based measurement in reading. The Journal of Special Education, 41 (2), 85 – 120. doi: 10.1177/00224669070410020401</bibtext> </blist> </ref> <ref id="AN0141337647-15"> <title> Footnotes </title> <blist> <bibtext> International Adult Literacy Survey.</bibtext> </blist> <blist> <bibtext> Adult Literacy and Life Skills Survey.</bibtext> </blist> <blist> <bibtext> Australian data are not available in the public use files.</bibtext> </blist> </ref> <aug> <p>By Egil Gabrielsen and John Sabatini</p> <p>Reported by Author; Author</p> </aug> <nolink nlid="nl1" bibid="bib12" firstref="ref1"></nolink> <nolink nlid="nl2" bibid="bib10" firstref="ref3"></nolink> <nolink nlid="nl3" bibid="bib14" firstref="ref5"></nolink> <nolink nlid="nl4" bibid="bib16" firstref="ref6"></nolink> <nolink nlid="nl5" bibid="bib21" firstref="ref15"></nolink> <nolink nlid="nl6" bibid="bib13" firstref="ref16"></nolink> <nolink nlid="nl7" bibid="bib20" firstref="ref19"></nolink> <nolink nlid="nl8" bibid="bib17" firstref="ref20"></nolink> <nolink nlid="nl9" bibid="bib22" firstref="ref29"></nolink> <nolink nlid="nl10" bibid="bib15" firstref="ref32"></nolink> <nolink nlid="nl11" bibid="bib19" firstref="ref35"></nolink> <nolink nlid="nl12" bibid="bib18" firstref="ref46"></nolink>
Header DbId: eric
DbLabel: ERIC
An: EJ1240868
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: A Closer Look at Low-Performing Adult Readers in the Scandinavian Countries
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Gabrielsen%2C+Egil%22">Gabrielsen, Egil</searchLink> (ORCID <externalLink term="http://orcid.org/0000-0003-4296-5591">0000-0003-4296-5591</externalLink>)<br /><searchLink fieldCode="AR" term="%22Sabatini%2C+John%22">Sabatini, John</searchLink> (ORCID <externalLink term="http://orcid.org/0000-0002-0292-2039">0000-0002-0292-2039</externalLink>)
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22Scandinavian+Journal+of+Educational+Research%22"><i>Scandinavian Journal of Educational Research</i></searchLink>. 2020 64(2):242-255.
– Name: Avail
  Label: Availability
  Group: Avail
  Data: Routledge. Available from: Taylor & Francis, Ltd. 530 Walnut Street Suite 850, Philadelphia, PA 19106. Tel: 800-354-1420; Tel: 215-625-8900; Fax: 215-207-0050; Web site: http://www.tandf.co.uk/journals
– Name: PeerReviewed
  Label: Peer Reviewed
  Group: SrcInfo
  Data: Y
– Name: Pages
  Label: Page Count
  Group: Src
  Data: 14
– Name: DatePubCY
  Label: Publication Date
  Group: Date
  Data: 2020
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: Journal Articles<br />Reports - Research
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Adult+Literacy%22">Adult Literacy</searchLink><br /><searchLink fieldCode="DE" term="%22Low+Achievement%22">Low Achievement</searchLink><br /><searchLink fieldCode="DE" term="%22Adults%22">Adults</searchLink><br /><searchLink fieldCode="DE" term="%22Reading+Skills%22">Reading Skills</searchLink><br /><searchLink fieldCode="DE" term="%22Reading+Tests%22">Reading Tests</searchLink><br /><searchLink fieldCode="DE" term="%22Foreign+Countries%22">Foreign Countries</searchLink><br /><searchLink fieldCode="DE" term="%22Sentences%22">Sentences</searchLink><br /><searchLink fieldCode="DE" term="%22Reading+Comprehension%22">Reading Comprehension</searchLink><br /><searchLink fieldCode="DE" term="%22Decoding+%28Reading%29%22">Decoding (Reading)</searchLink><br /><searchLink fieldCode="DE" term="%22Printed+Materials%22">Printed Materials</searchLink><br /><searchLink fieldCode="DE" term="%22Time+on+Task%22">Time on Task</searchLink><br /><searchLink fieldCode="DE" term="%22Basic+Vocabulary%22">Basic Vocabulary</searchLink><br /><searchLink fieldCode="DE" term="%22Scores%22">Scores</searchLink>
– Name: Subject
  Label: Geographic Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Denmark%22">Denmark</searchLink><br /><searchLink fieldCode="DE" term="%22Norway%22">Norway</searchLink><br /><searchLink fieldCode="DE" term="%22Sweden%22">Sweden</searchLink>
– Name: SubjectThesaurus
  Label: Assessment and Survey Identifiers
  Group: Su
  Data: <searchLink fieldCode="SU" term="%22Program+for+the+International+Assessment+of+Adult+Competencies+%28PIAAC%29%22">Program for the International Assessment of Adult Competencies (PIAAC)</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1080/00313831.2018.1539033
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 0031-3831
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: This article focuses on low-performing adult readers in the Scandinavian countries, seeking to identify and describe some of the literacy-related problems that they face. Following a presentation of the Reading Components Assessment (RCA) used in PIAAC (Program for International Assessment of Adult Competencies), we provide profiles for those informants who completed the RCA. Then we compare the RCA results of adults having well-functioning literacy skills with the results of adults identified as low-performing readers. The presentation includes both results for items correct and information about the time used to complete the items. The findings indicate that the gap in performance on the Component task sets is wider between adults scoring "Below level 1" in literacy and those scoring on Level 1 than the corresponding gap between "Level 1" and "Level 2." Implications for policy and practice are discussed.
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2020
– Name: AN
  Label: Accession Number
  Group: ID
  Data: EJ1240868
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1240868
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1080/00313831.2018.1539033
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 14
        StartPage: 242
    Subjects:
      – SubjectFull: Adult Literacy
        Type: general
      – SubjectFull: Low Achievement
        Type: general
      – SubjectFull: Adults
        Type: general
      – SubjectFull: Reading Skills
        Type: general
      – SubjectFull: Reading Tests
        Type: general
      – SubjectFull: Foreign Countries
        Type: general
      – SubjectFull: Sentences
        Type: general
      – SubjectFull: Reading Comprehension
        Type: general
      – SubjectFull: Decoding (Reading)
        Type: general
      – SubjectFull: Printed Materials
        Type: general
      – SubjectFull: Time on Task
        Type: general
      – SubjectFull: Basic Vocabulary
        Type: general
      – SubjectFull: Scores
        Type: general
      – SubjectFull: Denmark
        Type: general
      – SubjectFull: Norway
        Type: general
      – SubjectFull: Sweden
        Type: general
      – SubjectFull: Program for the International Assessment of Adult Competencies (PIAAC)
        Type: general
    Titles:
      – TitleFull: A Closer Look at Low-Performing Adult Readers in the Scandinavian Countries
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Gabrielsen, Egil
      – PersonEntity:
          Name:
            NameFull: Sabatini, John
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 01
              Type: published
              Y: 2020
          Identifiers:
            – Type: issn-print
              Value: 0031-3831
          Numbering:
            – Type: volume
              Value: 64
            – Type: issue
              Value: 2
          Titles:
            – TitleFull: Scandinavian Journal of Educational Research
              Type: main
ResultId 1