Reading in a Foreign Language
Volume 18, Number 2, October 2006
ISSN 1539-0578

shem

Toward a multicomponent model of interest and L2 reading: Sources of interest, perceived situational interest, and comprehension

Cindy Brantmeier
Washington University
United States

Abstract

With 104 advanced second language (L2) readers, this investigation seeks to construct a multicomponent model of interest and L2 reading. Sources of interest (SI) refer to variables that induce feelings of interest in a text. Perceived interest (PI) refers to the feeling of interest itself (Schraw, Bruning, & Svoboda, 1995). A factor analysis revealed the following 5 SI in order of variance: (1) cohesion (31%), (2) prior knowledge (13%), (3) engagement (12%), (4) ease of recollection (10%), and (5) emotiveness (6%).  PI accounted for 68% of the total sample variation. Of the 5 factors on the SI questionnaire, three were uniquely related to PI: cohesion, engagement, and ease of recollection. PI was related to sentence completion items and multiple choice items, but not recall. Five factors from the SI questionnaire were regressed on comprehension scores after first removing the variance explained by perceived interest. Ease of recollection positively related to all 3 comprehension tasks. Findings provide a construct of interest for L2 reading and serve as an attempt to develop an instrument to assess different sources of interest in L2 reading. A preliminary understanding of these dimensions of interest as predictors of L2 reading comprehension is presented.
keywords: Multicomponent model of interest, second language reading, comprehension

In a review article on new directions in second language (L2) reading research, Bernhardt (2003) specifically stated that the role of interest in L2 reading comprehension has yet to be fully explored. Additionally, she contended that examining ephemeral factors such as engagement, interest, and purpose with readers in high-stake situations will yield significant insights into the L2 reading process (p. 38). The present study relied on research in cognitive psychology and first language (L1) reading to examine the relationship among sources of interest, perceived interest, and comprehension in L2 reading. To date, few studies of L1 or L2 interest and comprehension have used multiple comprehension measures. To explore whether reader interest in an L2 text may be a function of purpose, readers in the present study completed a written recall task, sentence completion task, and multiple choice questions as part of an in-class examination after reading a lengthy, authentic text. The present study utilized readers from the advanced levels of language instruction in an attempt to fill the lacuna in the database of L2 research involving the advanced reader (Brantmeier, 2001; Young, 2003).

The purpose of the present study was to consider how one current conceptualization of interest in L1 reading with adults might inform understanding of interest in L2 reading with adults. Does the concept of interest in L2 reading become more like L1 reading as learners become more advanced? The present study utilized dimensions of interest identified by Schraw, Bruning, and Svoboda (1995) in order to examine whether these dimensions draw on a similar construct in L2 reading with advanced readers and to offer a preliminary understanding of these dimensions of interest as predictors of L2 reading comprehension. These predictions may offer insight into the unexplained variance in L2 reading by proposing assumptions and predictions about the relationship between the different sources of interest in the L2 reading context. More specifically, the present L2 study attempted to examine the relationships among sources of interest, perceived interest, and three different comprehension assessment tasks (i.e., written recall, sentence completion items, and multiple choice items).

L2 Reading Models

Berhnhardt's (1991) pioneering model combined both cognitive and social perspectives on reading as it included both text-driven and reader-based views of the L2 reading process. This integrative perspective assumed that reading development and reading proficiency exist. The model encompassed micro-level features, such as word recognition, phonemic/graphemic features and syntax, as well as macro-level features, such as background knowledge and perceptions (both knowledge-driven features) (Brantmeier, 2004). More recently, Bernhardt (2005) proposed a compensatory model of L2 reading that included unexplained variance as a factor. In this model, based on Stanovich's (1980) model, knowledge sources assist for other sources that are deficient or non-existent. The three-dimensional model included L1 literacy knowledge and language knowledge (emphasizing lexical items), which account for 50% of the variation in L2 reading, and it also incorporated dimensions yet to be explained, such as interest. This model was the first L2 reading model to illustrate that knowledge sources operate synchronically, interactively, and synergistically. It emphasized the L2 reading process as a "juggling" process in cognition (p. 140). The present study attempted to begin to conceptualize interest as a variable involved in the L2 reading process.1 Instead of trying to detect linear relationships between certain individual difference factors and corresponding outcome or performance variables in isolation, the study attempted to work with a more complex theoretical paradigm. Dimensions of a specific variable, interest, may have more predictive power than variables in isolation.

Interest and L1 Reading

With readers of all ages, L1 literacy researchers have examined several types of interest involved in the reading process and consequently have provided operational definitions of interest. In the most recent Handbook of Reading Research, Alexander and Jetton (2000) offered a detailed review of definitions and categories of interest (p. 298). They highlighted situational and personal interest as the two main dimensions of interest involved in the reading process. Individual or personal interest involves the readers' preferences for certain passage topics or subject matter, and this sustained interest exists before reading a particular text (Hidi, 1990; Schiefele, 1992). Situational interest refers to interest caused by situational variables, such as the text and test. This category of interest is short-lived and is induced by a particular situation (Krapp, Hidi, & Renninger, 1992; Wade, 1992). Hidi and Anderson (1992) defined text-based situational interest as interest that is evoked by text through topics or ideas that are of universal appeal.

Prior research on interest revealed that personal interest is connected to the learner's background knowledge or existing schema (Alexander, 1997) and intrinsic motivation (Deci, 1992; Schiefele, 1991) and that logically situational interest is aroused spontaneously (Hidi & Anderson, 1992). Both personal interest and situational interest have been investigated in the L1 reading process (Alexander, Kulikowich, & Schulze, 1994; Hidi, 1990; Schiefele, 1992), and results revealed that personal interest positively affects reading comprehension. Alexander, Jetton, and Kulikowich (1995) found that students with little domain or topic knowledge and who were generally uninterested in the domain and in the passages performed lower than their counterparts on a written recall task. Overall, researchers have asserted that a positive relationship exists between personal interest, prior knowledge, and comprehension.

Table 1: Review of research — factors affecting situational interest and reading comprehension with L1 readers
Authors Factors
Schank, 1979; Kintsch, 1980 Emotionally charged or provocative information
Anderson, Shirey, Wilson, & Fielding, 1987 Character identification
Grolnic & Ryan, 1987; Deci, 1992 Motivational constraints (such as internal locus of control)
Iran-Nejad, 1987 Unexpectedness of main events
Hidi, 1990 Topic shift
Schiefele, 1991 Value-related feelings
Garner, Brown, Sanders, & Menke, 1992 Isolated segments
Wade, 1992 Text cohesion
Shirey, 1992; Schraw & Dennison, 1994 Relevance of information to readers' goals
Sadowski, Goetz, & Fritz, 1993 Concreteness and imageability of salient text segments
Mitchell, 1993 Degree to which test information engages readers

Some studies have specifically examined the effects of situational interest on comprehension. Schraw et al. (1995) offered a succinct review of research, and Table 1 depicts the variety of factors involved (see Schraw et al., 1995, for a more detailed discussion of these studies). As indicated in the table, in the past decade research reported notable findings. Schraw and Dennison (1994) examined the effects of readers' purpose on comprehension and reported that text segments involving purpose-driven interest are recalled better than segments that are not relevant to readers' purpose. With college students enrolled in an introductory education course, Wade, Buxton and Kelly (1999) used multiple measures to examine what text characteristics readers found interesting and uninteresting and how interest affected recall in two expository texts. Findings revealed that the following five text characteristics were most associated with interest: (a) information that was important, new, and valued; (b) information that was unexpected; (c) connections readers made between the text and their prior knowledge or experience; (d) imagery and descriptive language; and (e) authors' connection (pp. 207-208). Regarding comprehension, readers more successfully recalled information that they rated as both interesting and important.

With university students enrolled in an introductory educational psychology course, Schraw et al. (1995) reported that different sources of interest (SI) affect perceived interest (PI) with subsequent effects on recall. The researchers created a questionnaire containing items from previous research (cited above) that included sources of situational interest variables and perceived situational interest variables to see whether SI affects PI and text recall. Sources of situational interest focused on the reader's assessments of the text's content and structure. See Table 2 for operational definitions of each subscale on the SI questionnaire.

Table 2: Six sources of situational interest
Dimension Operational Definition
Ease of comprehension Assesses whether the text was easy to remember and concentrate on.
Cohesion Pertains to the text's organization and clarity.
Vividness Addresses the degree to which the text contained exciting and vivid details.
Engagement Measures the degree to which the text was thought-provoking, stimulating, and timely.
Emotiveness Addresses whether the story evoked strong emotional responses.
Prior knowledge Measures the degree to which the reader was familiar with the story's content prior to reading.2
Schraw et al. (1995)

The PI questionnaire focused on the reader's assessment of his or her own feelings of interest and measured overall situational interest in the content and issues raised by the text. The researchers found that perceived interest correlated significantly with total recall, and when removing the variance explained by perceived interest, the following three variables explained 18% of the variance in recall: ease of comprehension, vividness, and engagement (Schraw et al., 1995, p. 10). When perceived interest was entered into the equation before sources of interest, the only variable related directly to recall was ease of comprehension. In the end, the authors claimed that the relationship between sources of interest and recall was mediated by perceived interest. In an attempt to create a preliminary multicomponent model of interest and L2 reading, the same perceived interest and sources of interest questionnaires were utilized in the present investigation to explore whether the above finding holds true with non-native speakers reading authentic, L2 texts. In addition to utilizing the written recall to measure comprehension (Schraw et al., 1995), in the present study readers also completed a sentence-completion task and multiple-choice questions.

L2 Research on Interest and Reading Comprehension

The pioneering work of social psychologists Gardner and Lambert (1972) showed that motivation and attitude affect general L2 learning. There are a plethora of hypotheses raised in the database of research about the connections of attitudes, motivations, and orientations to achievement in the L2 (see Masgoret & Gardner, 2003), but it appears to date that little research exists on the specific role of interest in L2 reading, although this issue is obviously linked to motivation as some studies on the effects of motivation have suggested that students' interest in learning an L2 is a significant factor in acquisition of language. Dornyei (2003) offered a review of research on motivation to learn a foreign language or L2 and included interest as a factor involved in situational conception of L2 motivation. More recently, Dornyei (2005) discussed the role of task-based activities in shaping learners interest. Keller (1984) connected interest to motivation and asserted that interest is a positive response to stimuli based on existing cognitive structures (background knowledge) in such a way that the learner's curiosity is aroused and sustained. Keller concluded that interest is one of the main elements of motivation in L2 learning.

To date, there is no conclusive evidence about the effects of interest on L2 reading comprehension. With English as L2 learners, Carrell and Wise (1998) examined the relationship between prior knowledge and topic interest in L2 reading and found no significant effects of the independent variables on multiple choice (MC) tests. They also reported that prior knowledge and topic interest were not highly correlated. The researchers utilized 10 MC items to measure prior knowledge. The interest inventory included a list of 10 topics where students reported their level of interest from 1 (most interest) to 10 (least interest). With intermediate level L2 learners of Spanish, Brantmeier (2003) investigated the effects of enjoyment and interest in L2 reading. She used a written recall assessment task to measure comprehension and found that although passage content and topic familiarity may increase the L2 readers' burden, enjoyment and interest mattered little at this level of instruction. For this study, topic familiarity, enjoyment and interest levels were assessed via multiple-choice questions (on a 5-point Likert scale) that allowed respondents to show discrimination in their judgments. The items on the questionnaire asked readers to rate how much they enjoyed reading the passages and how interested they were in the passages. Participants indicated low levels of enjoyment and interest but these factors did not hinder performance on written recalls. At this level enjoyment and interest did not predict comprehension. Brantmeier asserted that future studies could provide further evidence about the role of interest in L2 reading. The present study attempted to utilize questionnaires and evidence from prior L1 reading and cognitive psychology research to further examine the ephemeral variable of interest. Again, in order to gain a more complete depiction of L2 reading comprehension, the study utilized multiple assessment tasks to measure comprehension.

Comprehension Assessment Tasks

Alderson (2000) asserted that there is no one best method for testing reading. Some reading assessment measures include multiple choice items, written and oral recall, cloze, sentence completion items, open-ended question, true/false, matching activity, checklist, and fill-in-the-blank tests. Individual assessment tasks provide a limited representation of reading comprehension; but despite this well-known assertion, many reading researchers continued to use only one task to measure comprehension. A variety of assessment tasks are needed in order to understand the complete picture and to be able to generalize research findings (Bernhardt, 1991). Research has also shown that the relationships among the tasks should be analyzed in order to examine the validity of the reading comprehension tests (Bernhardt, 1991; Wolf, 1993). Some prior studies have examined how readers performed on different types of comprehension assessment tasks with different types of readings (Brantmeier, 2005; Carrell, 1991; Lee, 1990; Shohamy, 1982; Wolf, 1993). The present study used the following comprehension tasks: written recall, open-ended sentences, and multiple choice items.

Motivation for the Present Study

As previously stated, based on prior research and theory in L1 reading, Schraw et al. (1995) proposed a multicomponent model that distinguished six potential sources of situational interest. They developed a valid and reliable instrument to assess different dimensions of interest in reading, and they explored the link between these dimensions and comprehension. To measure comprehension, the researchers utilized the written recall and reported that PI was related to recall, and once this effect was removed from the regression equation, only ease of comprehension explained a significant proportion of variance in text recall. As discussed earlier, these findings may hold important conceptual implications for the current L2 reading models.

Research Questions

The following questions guide the present study:

  1. Are there factors that elicit situational interest during L2 reading?
  2. Is perceived interest related to L2 written recall, sentence completion items, and multiple choice items?
  3. Are sources of interest related to L2 written recall, sentence completion items, and multiple choice items, once the effect of perceived interest is controlled?

These questions are significant because to date it appears that no study has examined the dynamic interplay among variables that elicit situational interest in an L2 text and to what extent these factors affect L2 reading comprehension as measured via varied assessment tasks.

Methods and Procedures

Participants

Participants, aged 18-21, were enrolled in an advanced level Spanish grammar and composition course at a private university. In previous courses at the university all participants in the study had previously read short readings from newspaper articles, magazines and historical vignettes. In the advanced grammar course, students were assigned to read lengthy, authentic literary works from the literary canon for the first time. As part of the course requirements they read complete short stories from a literary anthology designed for this level of acquisition. One goal of the course was to prepare learners for the level of reading, writing, and speaking necessary to be successful in the advanced literature courses. At the university where data were collected there is no language requirement, and therefore all students in the study had enrolled in the course voluntarily.

To ensure a more homogeneous population of learners, only students with the following criteria were included in the final data analysis: students who were placed in the course based on scores from the national Advanced Placement Spanish exam and who achieved the appropriate placement score on the online departmental placement exam; or students who took prerequisite Spanish courses at the university. Furthermore, only students whose native language was English and only those who completed all tasks were included. In the end, 104 students (46 men and 58 women) were included in the final analysis.

Reading Passage

The reading passage was selected after carefully looking at different literary texts that are used at this level of language instruction. The short story Aniversario, by Luis Romero, was taken from an anthology entitled Aproximaciones al Estudio de la Literatura Hispánica, by Virgillo, Friedman, Valdivieso. The story consisted of 1, 218 words and was kept in its original form, including word glosses.3 A male adolescent who died years ago narrated the story. A family is sitting at the dinner table talking about daily activities. The father, mother, son and daughter discuss their plans for the evening, which include playing soccer and going to the movies. They talk about the recent happenings in the neighborhood. In the end the mother is upset because no one remembers that today is the anniversary of the death of his or her son/brother. The entire story takes place in the house at the dinner table. The entire passage is included in Appendix A.

A pilot study was conducted with 83 participants to determine if there were topic familiarity differences by gender with the reading passage. Findings revealed that male and female participants reported being equally familiar with the passage topic (males M = 3.44, SD = .68; females M = 3.40, SD = .86). An analysis of variance revealed no significant differences by gender for self-reported topic familiarity levels (p < .05).

Written Recall, Sentence Completion Tasks, and Multiple Choice Tasks

Prior studies have shown that readers perform differently on different types of comprehension assessment tasks (Bernhardt, 1991; Brantmeier, 2005; Carrell, 1991; Lee, 1990; Shohamy, 1982; Wolf, 1993), and consequently the present study used the following comprehension tasks: written recall, open-ended sentences, and multiple choice items.

A common method used to measure L2 comprehension is the written recall task (Barnett, 1988; Brantmeier, 2001; 2003; Carrell, 1983; Lee, 1986a; 1986b; Maxim, 2002; among others). Bernhardt (1991) claimed that conducting the free recall does not influence a reader's understanding of the text in any way. She contended that with multiple-choice or open-ended questions additional interaction exists among texts, reader, questioner, and among the questions. When students are required to write freely they are not limited by the pre-determined and created assessment tasks. In other words, the free-written task does not deny the role of the individual reader in meaning construction. In this study the written instructions on the recall page told the learner to try to recall main ideas, as well as details about the passage read. More specifically, the written recall task asked participants, without turning back to the passage, to write down in English as much as they could about the passage just read.

The sentence completion task uses retrieval cues, but it is an open-response task with certain limits placed on possible answers. The sentences are created so that all possible answers are foreseeable, and the objectivity of scoring depends on the comprehensiveness of the answer key. Multiple-choice questions, the most widely used means of assessing reading comprehension, include retrieval cues, and the answers are predetermined with no ambiguity in the scoring of right or wrong answers. The multiple-choice questions were created to meet the two criteria set by Wolf (1993): (a) All items are passage dependent, and (b) some of the items require the reader to make inferences. In addition, a third condition was included: Correct responses could not be determined by looking at the other questions on the page. For each of the 10 multiple-choice questions four possible responses were created: one correct response and three distractors. All distractors in the multiple-choice questions are plausible (Wolf, 1993), and all multiple-choice questions could not be answered correctly without having read and understood relevant parts of the passages. To test the multiple choice items, 45 native Spanish speakers completed the readings and assessment tasks. The sentence completion (SC) and MC questions were created so that all questions corresponded or had an equivalent information or inferential focus (Wolf, 1993). Two different reading specialists were consulted to ensure that the SC and MC questions corresponded. Furthermore, an internal consistency index was calculated for the sentence completion test and multiple choice items (KR-20 = .873). See Appendix B for examples of a SC task and a corresponding MC item.

Prior research has also shown that when the readers are allowed to use their native language in the comprehension assessment tasks, a truer depiction of comprehension is revealed. Therefore, all assessment tasks in this study were completed in the learner's native language, English (Bernhardt, 1983; Lee & Ballman, 1987; Wolf, 1993).

Sources of Interest Questionnaire (SIQ)

The SIQ questionnaire was taken from Schraw et al. (1995), and it contained 17 items in which individuals reported on a 5-point Likert scale the degree to which they agreed or disagreed with each statement about the text. See Table 3 to view the 17 items included on the SIQ questionnaire.4

Perceived Interest Questionnaire (PIQ)

The PIQ was taken from Schraw, Bruning, and Svoboda (1995), and it contained 10 items that were slightly modified according to the passage utilized for the present study. For each item readers indicated the degree to which they agreed or disagreed with each statement (see Table 4). This instrument assessed feelings of situational interest in the text immediately after reading.

Table 3: Items included in the sources of interest questionnaire*
Cohesion The information in the story was well organized. (.69)
The story was easy to understand. (.80)
The story's main ideas were presented clearly. (.80)
The story had a polished style. (.58)
Prior knowledge The story covered a topic I have read about or heard about before. (.86)
The story contained information I was familiar with.** (.88)
The story dealt with a topic I know a lot about. (.85)
Engagement The story was thought provoking. (.65)
The story dealt with highly relevant issues. (.59)
The story included vivid and exciting details. (.85)
Ease of recollection The story was easy to remember. (.57)
The story was easy to concentrate on. (.54)
The story was easy to picture in my head. (.59)
The story contained some unforgettable information. (.85)
Emotiveness The story made me happy. (.61)
The story made me upset.* (.81)
The story had certain eeriness to it. (.65)
* Numbers in parentheses are factor loadings.
** Slight changes were made to the item from Schraw et al. (1995) to reflect the nature of this passage topic and study.

Table 4: Items included in the perceived interest questionnaire*
I thought the story was very interesting. (.77)
I'd like to discuss this story with others at some point. (.90)
I would read this story again if I had the chance. (.83)
I got caught-up in the story without trying to. (.72)
I thought the story's topic was quite interesting.** (.62)
I think others would find this story interesting. (.60)
I would like to read more stories like this in the future.** (.68)
This story was one of the most interesting things I've read in Spanish in a long time. (.89)
This story really grabbed my attention. (.89)
*Numbers in parentheses are factor loadings.
** Slight changes were made to the item to reflect the nature of this topic and study; the item "I'll probably think about the implications of this story for some time to come" was omitted because it was not relevant to this narrative.

Topic Familiarity

In addition to the items in the SIQ and PIQ that dealt with familiarity of topic, another brief question was utilized. In this case, topic familiarity was assessed with a question that included five possible choices that ranged from "I knew a lot about the topic" to "I did not know anything about the topic at all." 5 The 5-point scale was used to encourage more precision in rating and encouraged respondents to show greater discrimination in their judgments.

Data Collection Procedures

In the middle of the semester during regular class time (50 minutes total) all participants were asked to complete the following: short story, SIQ questionnaire, PI questionnaire, written recall task, sentence completion items, multiple choice items, topic familiarity questionnaire. Participants were told that they would read a passage (no fixed time) and then complete comprehension assessment tasks. They were instructed not to look back at any previous pages while reading and completing all tasks. The researcher or research assistant was present at all data collection times to ensure that participants did not look back at the passage when completing the comprehension assessment tasks.

Scoring and Analysis Procedures

The short story was divided into pausal units6 by three different raters. Pausal units were defined as a unit that has a "pause on each end of it during normally paced oral reading," (Bernhardt, 1991, p. 208). Separately, the researcher and two additional raters identified the total pausal units for the text and then compared results. The percent of scoring agreement between the three raters was .96. In the end, the total number of pausal units was 133.

Sentence completion items were scored according to a template of correct and acceptable answers supplied by the researcher and a research assistant. Both sets of scores (written recalls and sentence completion scored by both the researcher and a research assistant) were submitted to correlational analysis to determine interrater reliability. Raters agreed on all scores for the sentence completion items. For recall, interrater reliability was .95. A third rater was consulted for the disagreement in recalls to reach a final decision. The total number of correct answers was used for the dependent variables, recall, sentence completion items, and multiple choice items, respectively.

Results

Factor Analysis of SIQ and PIQ

The first analysis was conducted in order to examine the factor structure and internal consistency of the Sources of Interest Questionnaire (SIQ) and Perceived Interest Questionnaire (PIQ) for L2 reading. As in Schraw, Bruning, and Svoboda (1995), the SIQ was analyzed using principal components analysis with varimax rotation. The varimax rotation maximizes the variance of the squared loadings for each factor, instead of maximizing variance of squared loadings for each variable (Kim & Mueller, 1978). See Tables 3 and 4 for factor loadings.

Overall, the indicators clustered highly on each factor with 17 indicators corresponding to 5 underlying factors in the SIQ. Five extracted factors accounted for 71% of the sample variation in the instrument. The first factor accounted for 31% of the variance and appeared to represent a "cohesiveness" dimension. The following items loaded highly (lambdas > .50), the story was well organized, the story was easy to understand, the main ideas were presented clearly, and the story had a polished style; the scale had an internal consistency of .77. The second factor accounted for 13% of the variance and appeared to represent "prior knowledge" features. The following items loaded highly (lambdas > .50), the story covered a topic I have read about or heard about before, the story contained information I was familiar with, and the story dealt with a topic I know a lot about; an average of these items yielded satisfactory internal reliability, alpha = .86. The third factor accounted for 12% of the variance and appeared to represent an "engagement" dimension. The following items loaded highly (lambdas > .50), the story included vivid and exciting details7, the story was thought provoking, and the story dealt with some relevant issues; an average of these items yielded satisfactory internal reliability, alpha = .76. The item "the story included some vivid and exciting details" loaded on to "vividness" in Schraw et al. (1995). For the present study, all three items related to "vividness" in Schraw et al. loaded on to other factors, and consequently "vividness" was not an autonomous factor in the present L2 study.

The fourth factor accounted for 10% of the variance, and it seemed to represent an "ease of recollection" dimension. The following items loaded highly (lambdas > .50), the story was easy to remember, the story was easy to concentrate on, the story was easy to picture in my head8, and the story contained some unforgettable information9; an average of these items yielded satisfactory internal reliability, alpha = .79. The fifth factor accounted for 6% of the variance, and it appeared to represent "emotiveness" features. The following items loaded highly (lambdas > .50), the story made me upset and the story had certain eeriness to it10; and the following had a negative loading, the story made me happy.

In summary, the factors for the SIQ in order of variance were: (1) cohesion (eigenvalue = 5.21, variance explained = 31%), (2) prior knowledge (eigenvalue = 2.17, variance explained = 13%), (3) engagement (eigenvalue = 1.98, variance explained = 12%), (4) ease of recollection (eigenvalue = 1.63, variance explained = 10%), and (5) emotiveness (eigenvalue = 1.02, variance explained = 6%). Factors loadings for each variable are given in Table 3. The internal consistencies of the composite indices derived from each factor using Cronbach's alpha were .77, .86, .76, .79, and .50, respectively.

The Relationship between SI and PI

Table 5 shows the correlations among each of these factors, a combined score from the PIQ, a total recall score, a sentence completion score, and a multiple choice score. Several factors in the SIQ were significantly intercorrelated: cohesion and engagement, cohesion and ease of recollection, prior knowledge and engagement, prior knowledge and ease of recollection, engagement and ease of recollection, and engagement and emotiveness. These correlations were important to understand because as the value of one item on the questionnaire increased, the value of the other item also tended to increase. Foremost in these pairwise relationships were cohesion and engagement and cohesion and ease of recollection, which indicates a strong association between these items. The higher the value of cohesion, the higher the value of engagement and ease of recollection.

Separate factor analysis was calculated on the PIQ using a varimax rotation. The analysis revealed that PI accounted for 68% of the total sample variation. Factor loadings for each of the 10 items on the PIQ are given in Table 4. The internal consistency of the PIQ using Cronbach's alpha reached .94 and .79.

Table 5: Correlations among variables used in the experiment
  C PK ENG ER EM PIQ TR SC MC
C __ .23 .43** .39** -.02 .59** -.01 .39** .31*
PK   __ .36** .30* .13 .33**  .10 .12 -.20
ENG     __ .36** .33** .70**  .08 .46** .31*
ER       __ .14 .47**  .50** .43** .40**
EM         __ .12  .23 .22 .08
PI           __  .10 .36** .25*
TR               __ .25* .37**
SC                __ .52**
MC                 __
Note. C = Cohesion, PK = Prior knowledge, ENG = Engagement, ER = Ease of recollection, Emotiveness = EM, PI = Perceived interest, TR = Text recall, SC = Sentence completion, MC = Multiple choice.
* p< .05; ** p< .01; n = 104.

As predicted based on Schraw et al. (1995) findings, the SIQ and PIQ indicated construct validity. Each factor included a high-loading marker variable that was representative of the construct it was intended to measure.

Composite scores were generated for each of the five SIQ factors by averaging the ratings for all items with a significant loading on each factor. Means and standard deviations for the six sources of interest are shown in Table 6.

In order to determine the unique relationship of the five composite SIQ scores to the composite PIQ score a hierarchical multiple regression equation was calculated. Order of entry was free to vary (i.e., entered simultaneously). Results indicated that of the five dimensions on the SIQ, three were related to PI. The three related dimensions were: cohesion (β = .23, t = 2.52, p < .05); engagement (β = .56, t = 5.95, p < .01); and, ease of recollection (β = .24, t = 2.74, p < .01). Given these findings, it can be said that SI predicts PI quite well, with cohesion entering the equation first followed by engagement and ease of recall. A second regression analysis was performed in which all possible interactions were included in the equation. The results were the same as the above.

These three dimensions accounted for 61 % of the variance in perceived interest. In contrast, factors such as prior knowledge and emotiveness did not explain any variation in perceived interest even though they explained meaningful proportion of variance in SIQ.

Table 5 lists correlations between SI and PI. As noted, most of the five factors were moderately correlated with the PI, which indicated there was a substantial amount of variation in the PIQ that could not be explained by the SIQ. Overall, this analysis echoes findings by Schraw et al. (1995), where the SIQ and PIQ were perceived as different psychological constructs: text variables that facilitate reader interest and the subjective experience of interest in a text. These authors also performed a second regression analysis where they entered both SIQ and PIQ in the same factor analysis. Like the present study, these two belonged to separate factors.

The Relationship between PI and Recall, SC, MC

Means and standard deviations for PI and the three comprehension assessment tasks are listed on Table 6. A simple regression analysis revealed that PI is not significantly correlated with total recall (R2 = .01, MS = 27.15, p = .43). 11 On the other hand, a regression analysis revealed that PI is significantly correlated with both sentence completion (R2 = .13, MS = 38.63, p = .00) and multiple choice (R2 = .06, MS = 9.53, p = .04). PI does predict performance on comprehension when assessed via multiple choice items and sentence completion items. In summary, then, analyses revealed a positive correlation between PIQ and both SC and MC, but there was no positive correlation between PIQ and recall (see Table 5).

Table 6: Means and standard deviations for sources of interest, perceived situational interest, and comprehension assessment tasks
  M SD
Sources of interest    
       Cohesion 2.9 .66
       Prior knowledge 2.6 .78
       Engagement 2.5 .78
       Ease of recollection 3.3 .69
       Emotiveness 3.2 .55
Perceived situational interest 2.5 .78
Comprehension    
       Total recall 12.1 6.6
       Sentence completion 6.9 2.2
       Multiple choice 7.7 1.6
Recall total score possible = 133; score range = 28, maximum = 30, minimum = 2.
Total possible score for SC and MC = 10. N = 104.

The Relationship between SI, PI, and TR, SC, and MC

Five factors from the SIQ were regressed on total recall, sentence completion items, and multiple choice items after first removing the variance explained by perceived interest. As expected given earlier findings, the only two factors that positively related to total recall were cohesion (β = .31, t = -2.21, p < .05) and ease of recollection (β = .35, t = 2.57, p < .01). For sentence completion, two factors positively related: engagement (β = .34, t = 2.24, p < .05) and ease of recollection (β = .30, t = 2.30, p < .05). For multiple choice, three factors positively related: prior knowledge (β = .42, t = -3.72, p < .01), engagement (β = .38, t = 2.30, p < .05), and ease of recollection (β = .44, t = 3.43, p < .01). To summarize, ease of recollection was positively related to all three comprehension tasks, which emphasizes the importance of this factor in L2 reading. Engagement was positively related to both sentence completion and multiple choice but not recall, which may indicate that engagement plays a role in L2 reading when measured via tasks that contain retrieval cues.

Discussion

1. What factors elicit situational interest during L2 reading?

Factor analysis reveal five potential sources of situational interest in L2 reading: cohesion, prior knowledge, engagement, ease of recollection, and emotiveness. Four of these sources of situational interest echo previous findings in L1 studies: cohesion, prior knowledge, engagement, and emotiveness (Hidi, 1990; Schiefele, 1991; Schraw et al., 1995). Ease of recollection is a dimension unique to the present L2 study, and it consists of the following items: the story was easy to remember, the story was easy to concentrate on, the story was easy to picture in my head, and the story contained some unforgettable information. These items load on a single factor in the present study, and in Schraw et al. these items were part of two separate dimensions: ease of comprehension and vividness. After consultation with other researchers, this new dimension was labeled ease of recollection instead of ease of comprehension because of the distinctive loadings. All three items related to vividness in Schraw et al. loaded on other factors, and consequently vividness is not an isolated factor in the present study as it connects well to other dimensions. In a close look at the text utilized for the present study, there are only a few vivid descriptions of people, places and events. Text effect may explain the absence of vividness as a factor in the present study, and it may not be related to L2 acquisition stages or issues. Future inquiries that include passages with vivid descriptions and details could further validate these five sources of situational interest in order to create a more informed conception of interest and L2 reading. The framework in the present study is empirically testable and serves as an attempt to develop an instrument to assess different sources of situated interest in L2 reading.

Of the five sources of situational interest, three were related to PI: cohesion, engagement, and ease of recollection. This finding partially echoes L1 reading research where cohesion, engagement, ease of comprehension, vividness, prior knowledge and emotiveness were all uniquely related to PI, and this substantiates the assertion that individual sources of interest affect perceived interest. If so, perhaps different sources of interest and perceived interest are important when reading different types of texts (Schiefele, 1992). Schraw et al. (1995) utilized a passage adapted from Time magazine that included both expository and narrative elements, and the present study used a narrative. Researchers in L1 studies (Englert & Hiebert, 1984; Hiebert, Englert, & Brennan, 1983; Mandler, 1978) and L2 studies (Brantmeier, 2005; Carrell, 1984a, 1984b, 1985; Hammadou, 1990, 2000) have investigated how different text types might lead to different results in comprehension. In a review of L1 reading research on text structure, Roller (1990) concluded that studies have produced contradictory results leaving text-structure researchers with no explanation for the conflicting findings. Perhaps text type is a factor that influences how a reader's assessment of text content and structure (SI) affects reader's assessment of feelings of interest (PI). When reading a narrative text, readers often visualize or form a mental representation of what they are reading, and Denis (1982) reported that L1 readers actually see what they are reading in their head. In the present study, under the dimension ease of recollection, readers overall indicate that they somewhat agreed that the "story was easy to picture in my head" and "the story was easy to remember." Ease of recollection is one of the dimensions that is uniquely associated with PI, and perhaps the narrative text plays a role in this connection. This finding is not surprising as committing information to memory has been a key factor in L1 reading for some time now (Alexander & Jetton, 2000, p. 286). Moreover, in the present study cohesion and engagement are both dimensions associated with PI, and both include items involving text quality (e.g., well-organized, presented clearly, thought-provoking, relevant issues, etc.).

In the present study, prior knowledge is correlated with PI, and this finding somewhat contradicts results reported by Carrell and Wise (1998) where prior knowledge and topic interest (as measured by an interest inventory) are uncorrelated (p. 267). Baldwin, Peleg-Bruckner, and McClintock (1985) found similar results to Carrell and Wise (1998) and explained that adults can be knowledgeable about a topic without having any real interest in the topic. This explanation does not explain findings of the present study with advanced, adult L2 readers where prior knowledge and PI are correlated. These results can be interpreted in several ways. Like the present study, Schraw and Dennison (1994) found that knowledge is tied to students' interests, and later, in a discussion about the influence of interest on text processing, Alexander and Jetton (2000) recognized the importance of knowledge and its connection to interests in the reading process. Future L2 reading research may provide more insights into the connection between PI and SI by utilizing more in-depth measures of both knowledge and interest. Like Schraw et al. (1995), the present study measures familiarity with a topic instead of actual prior knowledge. In order to replicate Schraw et al., the author of the present study utilizes the same terms and operational definitions. These findings lend support for the hypothesis that topic familiarity (not prior knowledge) is a significant factor in explaining variation in PI with advanced readers. Bernhardt's (1991) model defines background knowledge as the reader's existing knowledge. The present study does not actually measure readers' existing knowledge. Alexander, Kulikowich, and Jetton (1995) used 25-item domain knowledge tests along with interest ratings (completed for entire readings and at paragraph level) to examine relationships with recall. Data collection instruments that are criterion-referenced, like that of Alexander et al. (1995), may provide further insights into the findings of the presents study. Clearly, the present study reduces a large number of interest-related items in a meaningful way, and it begins to capture the essence of interest and L2 reading.

2. Is perceived interest related to L2 comprehension?

The PIQ measured overall situational interest that focuses exclusively on the reader's assessment of his or her own feeling of interest. Schraw et al. (1995) reported that perceived interest was related to recall, but the present L2 study contradicts these L1 findings where no positive relationship between PI and recall was found. In the present study, the relationship between PI and recall can be considered as weak or even nonexistent, and this is difficult to interpret. Overall situational interest in the content and issues raised by the text does not hinder or facilitate written recall. Prior L1 reading research has shown that reading comprehension and recall are enhanced when readers read texts related to their interests (Renninger, Hidi, & Krapp, 1992), but this finding does not hold true for the present study. The overall PI items do not positively relate to recall, however a future study should examine the relationship between individual PI questions and recall to further explore this issue.

It is important to note that the total number of pausal units is 133, and readers in the present study average 12 (min = 2; max = 30) with a range of 28. In summary, readers only recall 11% of the total possible pausal units. This result is surprising given that these readers are enrolled in an advanced language course. To avoid recall as a test of writing rather than reading, all readers wrote in their native language. The low number of units recalled cannot be explained because of an L2 writing barrier. In their study, Schraw et al. (1995) utilized an 800 word passage and suggested that future research should utilize longer expository and narrative texts to test the theory of situational interest. The length of the reading passage in the present study, approximately 1,218 words, could explain the low average number of units recalled. A shorter passage may produce different results; however, the present study utilizes a passage of customary length for this level of instruction. Furthermore, it is standard practice for L2 reading research to use the pausal unit rubric to codify written recalls. Future research might explore cognitive processes that mediate the effects of interest on L2 recall with advanced readers (Tobias, 1995). Given the above limitations and issues, it is too early to make principled assertions about the relationship between PI and written recall in L2 reading.

It appears that prior investigations that examine interest as a variable in reading do not use assessment tasks other than the recall. The present study utilizes both sentence completion items and multiple choice items as further dependent variables. The strength of the association between other comprehension assessment tasks (PI and SC, and PI and MC) is statistically significant, indicating that a relationship among these variables does exist. PI does predict performance on comprehension when assessed via SC and MC.

A stronger relationship exists between PI and SC, where 13% of the variance in SC is accounted for by PI, with 6% of the variance in MC accounted for by PI. The SC task utilizes retrieval cues but is also considered an open-response task, and MC also includes retrieval cues but leaves no room for any open responses. In the present study, more of the variance in PI can be explained by SC than MC, which may suggest that the reader's assessment of his or her own feelings of interest relates better with a task where students write a brief response. With SC, it is possible to interpret readers' responses to see if they have really understood, whereas on MC items readers may select the correct answer by eliminating others. All in all, the significance of the relation between perceived interest and method of assessment is an area worthy of more in-depth and sustained inquiry.

3. Are sources of interest related to L2 recall, SC, and MC, once the effect of perceived interest is controlled?

Findings in L1 studies have indicated a positive relationship between interest and recall (Garner et al., 1992; Hidi & Baird, 1988; Iran-Nejad, 1987; Schraw et al., 1995; Wade & Adams, 1990; Wade, Schraw, Buxton, & Hayes, 1993). Schraw et al. (1995) reported that cohesion, ease of comprehension, vividness, and emotiveness are all significantly correlated with written recall. Findings in the present study reveal no overall positive relationship between PI and recall. This finding could be explained in the following way: The readers in the present study were from university level courses at a private institution taking advanced Spanish courses. Perhaps these readers are more strategic readers and chunked ideas together instead of recalling specific details. Future inquires with L2 readers should examine this phenomenon.

Once any effect (positive or negative) of PI is controlled, two sources of interest emerge (cohesion and ease of recollection) and are significantly correlated with recall. The cohesion dimension involves items such as "the story was well organized" and "the main ideas were presented clearly." The main ideas of the narrative text in the present study are logically connected and related, so it is not surprising that this dimension would significantly correlate with recall, an extended production response task (Bachman & Palmer, 1996). The immediate recall task asks the students to write down everything they can remember from the text, and the ease of recollection dimension includes items such as "the story was easy to remember" and "the story contained some unforgettable information," which may explain this significant relationship.

The ease of recollection dimension was also positively related to both SC and MC, which indicates that this dimension is a very powerful factor involved in the process undertaken by advanced L2 readers. Engagement is positively connected to both SC and MC, which may suggest that relevant and thought provoking issues are important factors involved in these structured tasks. Overall, findings indicate how prevailing the ease of recollection dimension is with advanced L2 readers. Prior knowledge is not significantly related to all assessment tasks. This finding contradicts prior research on the role of background knowledge and L2 reading with readers from beginning and intermediate levels of language instruction (Brantmeier, 2003, 2005; Carrell, 1981, 1983; Hudson, 1982; James, 1987; Johnson, 1981) that has revealed that what students already know (the knowledge stored in their existing cognitive domain) significantly affects their comprehension of L2 reading materials. In the present study of advanced readers prior knowledge does not play a significant role, and this lends support to Bernhardt's model of L2 reading. As discussed earlier, these results could be due to the test used to measure "prior knowledge" because it is not a measure of what students already know.

Another important point to address is that PI and SI do include some measure of self-assessment of comprehension. For example, several items on the SI questionnaire, such as "the story was easy to understand" and "the story was easy to remember" may measure self-assessment of comprehension in addition to interest. Recently, Brantmeier (2006) found that self-assessment of L2 reading ability, as measured before and after reading via a five point scale, is not an accurate predictor variable for placement or subsequent performance in reading. She asserted that a more contextualized, criterion-referenced self-assessment instrument may be beneficial for the placement of advanced readers in the USA. The present study sheds some light on this issue as items in the SIQ instrument could be tested to see if they elicit the reader's assessment of comprehension. A future study could specifically address this question. Instrumentation of self-assessment as a factor in L2 reading is an area that merits further research as it could be quite beneficial to those in charge of language placement.

In another study, Schraw (1997) utilized a recognition test instead of recall and found that SI and PI factors did not reach significance when regressed on the test. He notes that previous studies indicated that higher levels of SI are related to better recall (p. 446). Given these contradictory findings, at this point assertions about interest in L2 reading with advanced readers cannot be made. Future research could investigate the strategies advanced readers utilize when completing different assessment tasks. Perhaps different tasks draw on different constructs of interest across levels of language instruction. 12

Findings of the present study with advanced readers support prior research on the value of interest in the L2 reading process. Instructors across all stages of acquisitions should consider readers' interest as a relevant factor when creating course materials. As part of the Extensive Reading (ER) approach, Day and Bamford (1998) advocated that students select what they want to read, as students are more likely to read material in which they are interested. The ER approach embraces student's interests in the reading process, and the present empirical investigation underscores the role of interest, even at the advanced levels of language instruction (see Bamford and Day, 2004, for a comprehensive discussion of extensive reading activities).

Conclusion

L2 reading is a multivariate process involving a variety of text and reader characteristics, and the present study indicates that the role of interest in this phenomenon should not be ignored. The present study with advanced readers reveals five sources of interest in L2 reading with three dimensions connected to reader's perceived interest. Ease of recollection appears to be the only dimension that is related to three different comprehension assessment tasks. Finally, findings of the present study identify sources of interest similar to L1 studies; however, results contradict the relationships among interest and comprehension. Overall, the present investigation is an attempt to commence a theory of personal and situational interest when reading an extended, authentic L2 text. The relationships among sources of interest and L2 comprehension need to be tested further before interest is included as a definite component of the unexplained variance of Bernhardt's (2005) compensatory model of L2 reading. A future investigation might expand upon the present findings by examining if dimensions of interest relate and influence L2 reading comprehension across stages of acquisition with varied passages. Finally, a goal of identifying an archetypal advanced L2 reader profile may not be unrealistic.

Notes:

1. The present investigation was conducted during regular class time, and therefore only 60 minutes were allowed for completion of all tasks. This investigation does not account for L1 reading skills or L2 proficiency levels. It is an attempt to begin to establish a conceptual framework of interest and L2 reading. Given the results of the present study, the author is currently conducting a follow-up experiment that accounts for both L1 reading skills and L2 proficiency in order to examine if interest may account for the 50% of variance unexplained in the most current L2 reading model (Bernhardt, 2005).

2. Schraw, Bruning, and Svoboda (1995) used the term "prior knowledge" to refer to the degree that a reader is familiar with subject matter. In order to replicate this prior study, the author utilized this same term. The author recognizes that topic familiarity may not be a measure of the construct of prior knowledge.

3. To construct a preliminary multicomponent model of interest and L2 reading, multiple measures of comprehension (recall, sentence completion items, and multiple choice items) were utilized along with one passage for the present study. A future study might use multiple passages along with one measure of comprehension for reliability of measurement. The author does not attempt to generalize on the basis of one single passage, but rather she wishes to begin to construct a preliminary model of interest with L2 readers.

4. The present study utilized the 17 items that loaded on Schraw et al. (1995) source of interest questionnaire. Schraw et al. created this instrument based on previous findings from L1 studies discussed in the literature review.

5. The author recognizes that topic familiarity is not a measure of prior knowledge as prior knowledge involves conceptual knowledge one has of the domain or topic (Alexander, Kulikowich, & Jetton, 1994).

6. This propositional analysis system of rating is based on breath groups, or pausal units. To determine pausal units, two different native speakers were asked to read the passage out loud.

7. A researcher of psychology who is familiar with this area of inquiry was consulted to ensure interpretation of the factors for the present study.

8. This item loaded on to the vividness factor in Schraw et al. (1995).

9. This item loaded on to the engagement factor in Schraw et al. (1995). Again, research in the field of psychology was consulted about factor loadings.

10 This item loaded on the vividness factor in Schraw et al. (1995).

11. Schraw et al. (1995) utilized an 800 word text that included both expository and narrative elements. The story was codified into propositions and prepositional modifiers, a rubric designed by Bovair and Kieras (1985). As is standard in L2 reading research, the present study parses the story into pausal units. MS = Mean Square.

12. Finally, L2 reading has begun to perceive gender as a factor worth considering as part of the cultural context of the L2 reader (Brantmeier, 2002, 2003; Bugel & Buunk, 1996; Carrell & Wise, 1998; Chavez, 2001; Schueller, 1999; Young & Oxford, 1997). In an L1 study with high school readers, Ainley, Hillman, and Hidi (2002) examined gender and interest processes in response to literary texts. Findings revealed that performance by gender is similar for high interest reading materials, but with low interest reading materials boys do less well (p. 425). Comprehension was measured via MC and recall. Future inquiries of this type could consider the effects of SI and PI with comprehension by gender.

References

Ainley, M., Hillman, K., & Hidi, S. (2002). Gender and interest processes in response to literary texts: Situational and individual interest. Learning and Instruction, 12, 411-428.

Alderson, J. C. (2000). Assessing reading. NY: Cambridge University Press.

Alexander, P. A. (1997). Mapping the multidimensional nature of domain learning: The interplay of cognitive, motivational, and strategic forces. In M. L. Maehr & P. R. Pintrich (Eds.), Advances in motivation and achievement: Vol. 10 (pp. 213-250). Greenwich, CT: JAI.

Alexander, P., & Jetton, T. (2000). Learning from a text: A multidimensional and developmental perspective. In M. L. Kamil, P. B. Mosenthal, P. D. Pearson, & R. Barr (Eds.), Handbook of reading research: Vol. 3 (pp. 285-310). Mahwah, NJ: Erlbaum.

Alexander, P. A., Jetton, T. L., & Kulikowich, J. M. (1995). The interrelationship of knowledge, interest, and recall: Assessing a model of domain learning. Journal of Educational Psychology, 87(4), 559-575.

Alexander, P., Kulikowich, J., & Jetton, T. (1994). The role of subject-matter knowledge and interest in the processing of linear and nonlinear texts. Review of Educational Research, 64(2), 201-252.

Alexander, P. A., Kulikowich, J. M., & Schulze, S. K. (1994). How subject matter knowledge affects recall and interest. American Educational Research Journal, 31, 313-337.

Anderson, R. C., Shirey, L. L., Wilson. P. T., & Fielding, L. G. (1987). Interestingness of children's reading material. In R. E. Snow & M. J. Farr (Eds.), Aptitude, learning and instruction: Vol. III. Cognitive and affective process analysis (pp. 297-299). Hillsdale, NJ: Erlbaum.

Bachman. L. F., & Palmer. A. (1996). Language testing in practice. Oxford: Oxford University Press.

Baldwin, R. S., Peleg-Bruckner, Z., & McClintock, A. H. (1985). Effects of topic interest and prior knowledge on reading comprehension. Reading Research Quarterly, 20(4), 497-505.

Bamford, J., & Day, R. (Eds.) (2004). Extensive reading activities for teaching languages. New York, NY: Cambridge University Press.

Barnett, M. A. (1988). Reading through context: How real and perceived strategy use affects L2 comprehension. Modern Language Journal, 72, 150-160.

Bernhardt, E. B. (1983). Syntactic and lexical /semantic skill in foreign language reading comprehension: The immediate recall protocol. Die Unterrichtspraxis, 16, 27-33.

Bernhardt, E. B. (1991). Reading development in a second language. Norwood, NJ: Ablex.

Bernhardt, E. B. (2003). New directions in reading research: Second language perspectives. Reading Research Quarterly, 37(4), 112-117.

Bernhardt, E. B. (2005). Progress and procrastination in second language reading. Annual Review of Applied Linguistics, 25, 133-150.

Boviar, S., & Kieras, D. E. (1985). A guide to propositional analysis for research on technical prose. In B. K. Britton & J. B. Black (Eds.), Understanding expository text (pp. 315-362). Hillsdale, N.J: Erlbaum.

Brantmeier, C. (2001). Second language reading research on passage content gender: Challenges for the intermediate level curriculum. Foreign Language Annals, 34, 325-333.

Brantmeier, C. (2002). The effects of passage content on second language reading comprehension by gender across instruction levels. In J. H. Sullivan (Ed.), Research in second language learning: Literacy and the second language learner. Greenwich, CT: Information Age Publishing.

Brantmeier, C. (2003). Does gender make a difference? Passage content and comprehension in second language reading. Reading in a Foreign Language, 15, 1-27.

Brantmeier, C. (2004). Building a comprehensive theory of adult foreign language reading: A variety of variables and research methods. The Southern Journal of Linguistics [C. Brantmeier (Ed.), Adult foreign language reading: Theory, research, and implications. Charlotte, NC: The University of North Carolina at Charlotte], 27(1), 1-7.

Brantmeier, C. (2005). Effects of reader's knowledge, text type, and test type on L1 and L2 reading comprehension in Spanish. The Modern Language Journal, 89(1), 37-53.

Brantmeier, C. (2006). Advanced L2 learners and reading placement: Self-assessment, computer-based testing, and subsequent performance. System, 34(1), 15-35.

Bügel, K., & Buunk, B. P. (1996). Sex differences in foreign language text comprehension: The role of interests and prior knowledge. Modern Language Journal, 80, 15-31.

Carrell, P. L. (1981). Culture-specific schemata in L2 comprehension.  In R. Orem & J. F. Haskell (Eds.), Selected papers from the ninth Illinois TESOL/BE (pp. 123-132).  Chicago: Illinois TESOL/BE.

Carrell, P. L. (1983). Some issues in studying the role of schemata, or background knowledge, in second language comprehension. Reading in a Foreign Language, 1 (2), 81-92.

Carrell, P. L. (1984a). The effects of rhetorical organization on ESL readers. TESOL Quarterly, 18, 441-469.

Carrell, P. L. (1984b). Evidence of a formal schemata in second language comprehension. Language Learning, 34, 87-112.

Carrell. P. L. (1985). Facilitating ESL reading by teaching text structure. TESOL Quarterly, 19, 727-752.

Carrell, P. L. (1991). Second language reading: Reading ability or language proficiency. Applied Linguistics, 12, 159-79.

Carrell, P. L., & Wise, T.E. (1998). The relationship between prior knowledge and topic interest in second language reading. Studies in Second Language Acquisition, 20, 285-309.

Chavez, M.  (2001). Gender in the language classroom.  New York: McGraw Hill.

Day, R., & Bamford, J. (1998). Extensive reading in the second language classroom. New York, NY: Cambridge University Press.

Deci, E. L. (1992). The relation of interest to the motivation of behavior: A self-determination theory perspective. In A. Renninger, S. Hidi, & A. Krapp (Eds.), The Role of Interest in Learning and Development (pp. 43-70). Hillsdale, NJ: Erlbaum.

Denis, M. (1982). Imaging while reading text: A study of individual differences. Memory and Cognition, 10(6), 540-545.

Dornyei, Z. (2003). Attitudes, orientations, and motivations in language learning: Advances in theory, research and applications. Language Learning, 53(1), 3-32.

Dornyei, Z. (2005). The psychology of the language learner. New Jersey: Lawrence Erlbaum Associates.

Englert, C. S., & Hiebert, E. H. (1984). Children's developing awareness of text structures in expository materials. Journal of Educational Psychology, 76, 394-401.

Gardner, R. C., & Lambert, W. (1972). Attitudes and motivation in second language learning. Rowley, MA: Newbury House.

Garner, R., Brown, R., Sanders, S., & Menke, D. (1992). Seductive details and learning from a text. In A. Renninger, S. Hidi, & A. Krapp (Eds.), The role of interest in learning and development (pp. 239-254). Hillsdale, NJ: Erlbaum.

Grolnic, W. S., & Ryan, R. M. (1987). Autonomy in children's learning: An experimental and individual different investigation. Journal of Personality and Social Psychology, 52, 890-898.

Hammadou, J. (1990). The effects of analogy on French reading comprehension. The French Review, 64(2), 239-252.

Hammadou, J. (2000). The impact of analogy and content knowledge on reading comprehension: What helps, what hurts. The Modern Language Journal, 84, 38-50.

Hiebert, E. H., Englert, C. S., & Brennan, S. (1983). Awareness of text structure in recognition and production of expository discourse. Journal of Reading Behavior, 15, 63-79.

Hidi, S., & Baird, W. (1988). Strategies for increasing text-based interest and students' recall of expository text. Reading Research Quarterly, 23, 465-483.

Hidi, S. (1990). Interest and its contributions as a mental resource for learning. Review of Educational Research, 60, 549-572.

Hidi, S., & Anderson, V. (1992). Situational interest and its impact on reading and expository writing. In A. Renninger, S. Hidi, & A. Krapp (Eds.), The role of interest in learning and development (pp. 215-238). Hillsdale, NJ: Erlbaum.

Hudson, T. (1982). The effects of induced schemata on the short-circuit in L2 reading: Non-decoding factors in L2 reading performance. Language Learning, 32, 1-32.

Iran-Nejad, A. (1987). Cognitive and affective causes of interest and liking. Journal of Educational Psychology, 79, 120-130.

James, D. (1987). Bypassing the traditional leadership: Who's minding the store? Profession, 41-53.

Johnson, P. (1981). Effects on reading comprehension of language complexity and cultural background of a text. TESOL Quarterly, 15, 169-181.

Keller, J. (1984). Motivational design of instruction. In C. Reigeluth (Ed.), Instructional design theories and models (pp. 383-434). Hillsdale, NJ: Lawrence Erlbaum.

Kim, J., & Mueller, C. (1978). Factor analysis: Statistical methods and practical issues. Beverly Hills, CA: Sage Publications.

Kintsch, W. (1980). Learning from text, levels of comprehension or: Why anyone would read a story anyway. Poetics, 9, 87-89.

Krapp, A., Hidi, S., & Renninger, K. A. (1992). Interest, learning and development. In A. Renninger, S. Hidi, & A. Krapp (Eds.), The role of interest in learning and development (pp. 3-25). Hillsdale, NJ: Erlbaum.

Lee, J. F. (1986a). On the use of the recall task to measure L2 reading comprehension. Studies in Second Language Acquisition, 8, 201-12.

Lee, J. F. (1986b). Background knowledge and L2 reading. Modern Language Journal, 70, 350-354.

Lee, J. F. (1990). A review of empirical comparisons of nonnative reading behaviors across stages of language development. In H. Burmeister & P. Rounds (Eds.), Variability in second language acquisition (pp. 453-72). Eugene, OR: University of Oregon Press.

Lee, J. F., & Ballman, T. L. (1987) Learners' ability to recall and rate important ideas of an expository text. In B. Van Patten, T. R. Dvorack, & J. F Lee (Eds.), Foreign language learning: A research perspective (pp. 108-17). Rowley, MA: Newbury House.

Mandler, J. M. (1978). A code in the node: The use of story schemata in retrieval. Discourse Processes, 1, 14-35.

Masgoret, A. M., & Gardner, R. C. (2003). Attitudes, motivation, and second language learning: A meta-analysis of studies conducted by Gardner and associates. Language Learning, 53(1), 123-163.

Maxim, H. (2002). A study into the feasibility and effects of reading extended authentic discourse in the beginning German language classroom. Modern Language Journal, 86, 20-35.

Mitchell, M. (1993). Situational interest: Its multifaceted structure in the secondary school mathematics classroom. Journal of Educational Psychology, 85, 424-436.

Renninger, K. A., Hidi, S., & Krapp, A. (Eds.). (1992). The role of interest in learning and development. Hillsdale, NJ: Erlbaum.

Roller, C. M. (1990). The interaction of knowledge and structure variables in the processing of expository prose. Reading Research Quarterly, 25, 79-89.

Sadowski, M., Goetz, E. T., Fritz, J. B. (1993). Impact of concreteness on comprehensibility, interest, and memory for text: Implications for dual coding theory and text design. Journal of Educational Psychology, 85, 291-304.

Schank, R. C. (1979). Interestingness: Controlling Influences. Artificial Intelligence, 12, 273-297.

Schiefele, U. (1991). Interest, learning, and motivation. Educational Psychologist, 26, 299-324.

Schiefele, U. (1992). Topic interest and levels of text comprehension. In A. Renninger, S. Hidi, & A. Krapp (Eds.), The role of interest in learning and development (pp. 151-182). Hillsdale, NJ: Erlbaum.

Schraw, G. (1997). Situational interest in literary text. Contemporary Educational Psychology, 22, 436-456.

Schraw, G, & Dennison, R. S. (1994). The effect of reader purpose on interest and recall. Journal of Reading Behavior, 26, 1-18.

Schraw, G., Bruning, R., & Svoboda, C. (1995). Sources of situational interest. Journal of Reading Behavior, 27(1), 1-17.

Schueller, J. (1999).  The effect of two types of strategy training on foreign language reading comprehension: An analysis by gender and proficiency. Unpublished doctoral dissertation, University of Wisconsin, Madison.

Shirey, L. (1992). Importance, interest, and selective attention. In A. Renninger, S. Hidi, & A. Krapp (Eds.), The role of interest in learning and development (pp. 281-296). Hillsdale, NJ: Erlbaum.

Shohamy, E. (1982). Affective considerations in language testing. Modern Language Journal, 66, 13-17.

Stanovich, K. E. (1980). Toward an interactive-compensatory model of individual differences in the acquisition of literacy. Reading Research Quarterly, 16, 32-71.

Tobias, S. (1995). Interest and metacognitive word knowledge. Journal of Educational Psychology, 87(3), 399-405.

Wade, S. E., & Adams, B. (1990). Effects of importance and interest on recall of biographical test. Journal of Reading Behavior, 22, 331-353.

Wade, S. E. (1992). How interest affects learning from a text. In A. Renninger, S. Hidi, & A. Krapp (Eds.), The role of interest in learning and development (pp. 281-296). Hillsdale, NJ: Erlbaum.

Wade, S. E., Schraw, G., Buxton, W.M., & Hayes, M.T. (1993). Seduction of the strategic reader: Effects of interest on strategies and recall. Reading Research Quarterly, 28, 3-24.

Wade, S. E., Buxton, W.M., & Kelly, M. (1999). Using think-alouds to examine reader-text interest. Reading Research Quarterly, 34, 194-216.

Wolf, D. (1993). A comparison of assessment tasks used to measure FL reading comprehension. Modern Language Journal, 77, 473-89.

Young, D. J., & Oxford, R. (1997). A gender-related analysis of strategies used to process input in the native language and a foreign language. Applied Language Learning, 8, 43-73.

Young, D. J. (2003). Workshop on overcoming anxiety in the foreign language classroom. Language learning series. Washington University in St. Louis.

Appendix A

Brantmeier's reading material 1
Brantmeier's reading material 2
Brantmeier's reading material 3

Appendix B

Example of Sentence Completion Item

1. Today the mother feels _______________________

Example of Multiple Choice Item

2. Today, the mother is

a. Sad
b. Happy
c. Grouchy
d. Indifferent

Example of Sentence Completion Item

3. Later in the afternoon Lola plans to _____________________

Example of Multiple Choice Item

4. What does Lola have planned for the afternoon?

a. Go to a movie
b. Go to the park
c. Go shopping
d. Go to the library

About the Author

Cindy Brantmeier is Assistant Professor of Applied Linguistics and Spanish at Washington University in St. Louis. She is Co-Director of the Graduate Certificate in Language Instruction and Director of the Program in English Language Instruction. She oversees language program assessment and placement. Dr. Brantmeier's research interests include interacting variables in second language reading, reading research methodology, testing and assessment, and other related areas. She has given invited presentations of her research in Malaysia, Argentina, Costa Rica, Amsterdam, the Republic of Georgia, Oxford (UK), and elsewhere.

About RFL | Table of Contents | Past Issues | Subscribe | Editorial Board | Submissions | Contact RFL