A Comparison of Skilled and Less Skilled Student Writers' Summaries

Summarizing is an instructional technique frequently observed in the pedagogical and assessment practices of teachers across levels. It is used as a classroom practice to assess a learner’s reading and writing competence. Generally, summarizing promotes understanding of the texts, helps assess comprehension and engages students in active reading and writing tasks. A good summary simplifies concepts to help students understand constructs at the independent level; that is, instruction from the teacher is not necessary for the students to comprehend the text.

Summarizing, however, is not a simple task. According to Anderson (1999), it is a complex task that requires considerable reading and writing skills; hence, how a student summarizes may be an evidence of his/her skills as a reader and writer. In summarizing, a student is tasked to state the main points or the most important details of the text. He/she is expected to use his/her own words to create a condensed version of the original text.

Most of the studies on summarizing before 1980 did not show a comparative advantage for summarizing over simple rereading on measures of either comprehension or recall (Arnold, 1942; Germane, 1921a, 1921b; Howe and Singer, 1975; Stordahl and Christenson, 1956 in Vacca and Vacca, 1999). Analysis of these early studies indicated that summarization consisted of low-level multiple-choice tests of textual information, hardly the type of information that is likely to be encoded during summary writing. The instrument design of these studies did not help researchers in gaining knowledge on the processing of the text performed by students.

Nevertheless, the methodological constraints in those studies were addressed in the 80’s; extensive studies on summarization were undertaken during this period. These studies showed more positive methodology and results. A case in point was the Reder and Anderson’s(1980) study on the construction of summary passages. The students were prompted to write summaries of passages to convey the main points and eliminate the peripheral details. They found out that students understand better the summarized passages than the passages in the original form.

Other studies on summarizing focused on improving the summarization skills of the students. Day (1980) undertook a study with low-ability community college students as participants. She looked into the effectiveness of summarization training with and without explicit cuing intended to facilitate self monitoring. The results showed that providing students with rules for summarizing influenced their summarizing competence― detecting main ideas and deleting trivial information. The training profited all the students, although those students with writing problems seemed to need more training. The result of this study only showed that one’s benefit on training wouldbe dependent on his/her learning needs and capabilities. In the conclusion of her study, Day stressed that the training resulted in more consistent and accurate application of rules and improvement in quality of the summaries with less explicit instruction. The training, however, benefited more the advanced students. She recommended, though, that students with more severelearning problems should require a more explicit training.

Cunningham (1982) as cited in Barr et al., 1996, as regards effectiveness of the procedure on summarization in improving fourth-grade students’ ability to write one-sentence summaries of paragraphs, compared the summarization procedure to a placebo. This procedure differed from other instructional methods for teaching summarization because there was not an explicit rule taught. Rather, the students were led through a systematic procedure for developing summaries designed to enable them to correctly apply the rules for summarization. In Cunningham’s study, after nine days of treatment, the experimental subjects performed better than the control subject at composing one-sentence summaries.

In a similar strain, Taylor and her colleagues (Taylor, 1982; Taylor and Beach, 1984; Taylor and Berkowitz, 1980 as cited in Barr et al., 1996) examined the usefulness of engaging middle-grade students in the process of building hierarchical summaries of textbook material. In these studies, the teachers, through modeling, trained the students how to turn the main and the sub-headings into verbal summaries. In the independent practice, the students worked on their own summaries. Findings showed that the hierarchical training helped the students to recall information better than other studying strategies such as answering questions or additionalstudy.

There were also studies that investigated and compared the summarizing abilities of students. Brown and Day (1983), as cited in Barr et al., 1996, studied the summaries of older and more expert readers and the younger and less skilled readers. They found out that the older and the more expert readers’ summaries were better than their younger and less skilled reader counterparts’. Moreover, theirstudy revealed that the older readers planned their summaries more than what the younger and less skilled readers did.

In another study, Brown and Day (1983), as cited in Barr et al., 1996, examined the ability of children from fifth, seventh and tenth grades as well as the ability of college students to summarize. The participants were asked to apply the rules for summarization. Brown and Day used especially constructed tests to enable the participants to predict when each rule should be applied or at least would be applied by experts (college rhetoric teachers). Results showed that the first two rules (on the deletion of trivial and redundant information) was used with at least 90% accuracy by the three groups but the use of rules 3 and 5 was found problematic. Rule 3 was on giving superordinate terms for a category and rule 5 called for the writing or invention of a missing topic sentence.

Similarly, Rinehart, Stahl and Ericson (1986), as cited in Barr et al., 1996, applied Brown and Day’s rules for summarization. The study showed that the trained students focus attention on important information. On the other hand, McNeil and Donant’s (1982) study for summarization (cited in Barr et al., 1996) utilized the rules similar to van Dijk and Kintch’s work (1978) with grade five students. The training group received instruction on how to apply the rules to simple (third-grade level) passages. Effects were found favoring the training group on both the summary-writing activity for a new passage and the comprehension test for that passage.

Day (1986), as cited in Barr et al., 1996, demonstrated the importanceof metacognitive skills in summarizing. With the average and below-average writers in junior college as participants, Day provided training on summarization. The study revealed that those participants who received both summarization and self-management instructions showed the most improvement in strategy use. Furthermore, the higher-level students profited more from the instruction than the lower-level writers, suggesting that more explicit training in complex rules may be necessary to some students. Self monitoring was most effective when the rules were of moderate difficulty. Day suggested that until strategies become routine, students may be unable to use and monitor them simultaneously.

Other studies on summarization in the 1980’s showed the following findings: improved comprehension transfer to summarizing new texts (Bean and Steenwyk, 1984; Day, 1980; McNeil and Donant, 1982 as cited in Barr et al., 1996), increased summarizing efficiency (Hare and Borchardt, 1984 as cited in Barr et al., 1996), and increased recall (Doctorow et al., 1978; Linden and Wittrock, 1981 as cited in Barr et al., 1996) and improved test scores (Bean and Steenwyk, 1984 as cited in Barr et al., 1996).

Studies on summarization in the 90’s supported the value of writing summaries as informational writing task (Alvermann and Phelps, 1998; Brozo and Simpson, 1995; Vacca and Vacca, 1999). These studies suggest that summarization could lead to a better understanding of the text. Likewise, the more skilled readers and writers summarized better than the less skilled readers and student-writers could be trained for effective summarization. Hence, summarizing, as research showed, couldcontribute in developing and in enhancing students’ reading and writing abilities.

As shown in some researches (Day, 1980; Brown and Day, 1983 as cited in Barr et al., 1996), students observed rules for summarization and these rules helped them to construct better summaries. Brown and Day (1983), as cited in Barr et al., 1996, suggested five basic rules for summarization. The rules are the following: 

1. Delete material that is trivial
2. Delete material that is redundant
3. Provide a superordinate term or event for a list items or actions or members category
4. Select topic sentences
5. Invent topic sentences when missing

Based on the analysis of Brown and Day (1983) of effective summaries (cited in Barr et al., 1996), an effective summary entails the dropping of less important details and the focusing on the most important aspect of the text. Hence, the insignificant details in the text are excluded from the summaries. These rules were used by experts in summarizing texts (Brown and Day, 1983 as cited in Barr et al., 1996). Similar guidelines, the macrorules, were developed by van Dijk and Kintsch (1978). There were basic operations involved in comprehending and in remembering. The van Dijk and Kintsch’s macrorules were based on their studies of people who summarize effectively.

The macrorules are the following:
1. Include no unnecessary details. This is distinguishing important from trivial and repetitious information.
2. Collapse lists. In this step, the students categorize bits of information into larger categories.
3. Identify and develop topic sentences. Students search the text for explicit topicsentences and create their own sentences where none exists in a text.
4. Integrate information. Students use key concept, words, phrases and topic sentences to compose the summary.

Comparing the standards for summarizing set by Brown and Day and van Djik and Kintch, the rules were of the same nature and concern. What appeared to be different was the manner of presentation and the aspect for emphasis. Van Dijk and Kintch put together the rules related to the identification and the statement of topic sentence while Brown and Day separated theminto two different rules. Nevertheless, the rules in the two frameworks pointed to the same direction. Hence, a student-writer performs similar tasks regardless of the framework used. But for this research, the basis for evaluating the student-writers’ outputs was grounded on Brown and Day’s rules in summarizing. This framework was selected because it is more recent compared with van Dijk and Kintch’s framework and the scoring is more manageable. In the analysis of the summaries, what seemed to be apparent was that the participants were monolingual speakers of English. An investigation on how bilinguals summarize texts remains to be a fertile ground for research and there are numerous facets of summarizing left untouched in the literature like relating summarizing with other constructs. Hence, there are still gaps as far as summarizing is concerned most especially in the aspect of how bilinguals summarize in a second language. This study was an attempt to explore this minimally scathed ground.

This study explored the similarities and the differences in the summaries of the skilled and the less skilled student-writers who have English either as a second or as a third language and learned the English through formal schooling.

The specific questions dealt with in this study are the following:

1. How do skilled student-writers summarize a cause/effect text?
2. How do unskilled student-writers summarize a cause/effect text?
3. Is there a difference between the summaries of skilled student-writers and less skilled student-writers?