writing and its instruction. We begin by introducing the major features of a summary,

Chapter Two

Literature Review

This chapter presents a brief review of previous research related to summary

writing and its instruction. We begin by introducing the major features of a summary,

followed by a short discussion on how summaries are classified. Then we present

previous research on the necessity and the beneficial effects of summary writing

instruction. The procedures and two major approaches to summary writing instruction

are then discussed, followed by a brief section on the criteria for assessing a summary.

2.1 The Major Features of a Summary

Many researchers have provided various definitions regarding what constitutes

a summary, but there are some shared features. First, as Friend (2001) and Fitzgerald

& McBeth (1991) point out in their respective studies, a summary should be written in

paraphrase. That is to say, students need to put the material they have read into their

own words (Casazza, 1993). Second, several researchers have also pointed to the

important feature that a summary must contain the gist of the original passage (Friend,

2001; Fitzgerald & McBeth, 1991; Donn, 1988; Smith & Dahl, 1984; Seale, 1978).

Therefore, students need to demonstrate their ability to condense when they are

required to write a summary (Enos, 1988; Brown, Day, & Jones, 1983; Lardner &

Lundberg, 2001). Third, Fitzgerald & McBeth (1991) and Lardner & Lundberg (2001)

have emphasized that a summary should not include any personal reactions or ideas.

Thus, students need to learn how to restate the writers’ ideas accurately (Lardner &

Lundberg, 2001) and omit personal opinion (Casazza, 1993). Fourth, with respect to

the length of a summary, the consensus is that a summary should be shorter than the

original text (Friend, 2001; Casazza, 1993). Fitzgerald & McBeth (1991) further

suggest that a summary should be “no more than one quarter its original length”

(p. 146). However, for Barnet et al. (1988), the length of a summary could vary based

on the purpose and the need of the audience. To sum up, summary writing is a very

active process (Enos, 1988) and it can force students to analyze, evaluate and

synthesize the passage they have read (DeNight, 1992).

2.2 Classifications of Summaries

Summaries have been classified in different ways by different researchers. Hill

(1991) and Behan & DeWitt (2003), for example, classify summary writing into two

major types. One is written for the writer himself or herself and is generally referred

to as writer-based summaries. This particular type of summary helps the writer to

remember or organize the ideas. The other type is written for an external audience and

is referred to as reader-based summaries.

However, Arapoff (1970) classifies summaries in a different way. For him, there

are four types of summaries: enumerative, selective, referential, and comparative.

Arapoff further points out that many reports include enumerative parts which are

usually marked by connectors such as “First, Second, or finally.” As for selective

summaries, Arapoff sees them as most suited for reports with a climactic order. For

Arapoff, writing a referential summary could be the most difficult among the four

types because a referential summary usually contains many referential relationships

and many levels. It is unfortunate that no clear definition is provided in Arapoff (1970)

for comparative summaries but he does caution us that none of the four ways of

summarizing is “pure.”

Another scheme of classification is found in Bleck (2003), with three major

types: précis, descriptive summary, and analytical summary. A précis, according to

Bleck, is a kind of restatement. A descriptive summary, on the other hand, contains

not only all the information in the restatement, but also something beyond the

restatement. In other words, a descriptive summary contains more details than the

précis. According to Bleck, an analytical summary goes a step further, providing not

only all the relevant information included in a descriptive summary, but also

comments and critiques that go far beyond the given passage. That is, an analytical

summary is like an essay based on but not confined to the original passage.

The present study follows the classification proposed by Hill (1991) and Behan

& DeWitt (2003), and the summaries the subjects are required to write are essentially

reader-based summaries, for the reason that the subjects do not write the summaries

for themselves; on the other hand, they write the summaries for an external audience,

the graders. Moreover, if we examine closely the nature of the summaries in the study

on the basis of the classification scheme recommended by Bleck (2003), the

summaries naturally belong to the genre of a précis, because the subjects need only to

restate the main gist; they do not need to give additional comments or go beyond the

précis.

2.3 The Necessity of Summary Writing Instruction

As correctly observed by Hill (1991) and Sieben & Anthony (1982), although

summarization is an essential skill to college students, the instruction of summary

writing has rarely been emphasized. Brown, Day & Jones (1983) also point out that

skills of summary writing cannot be automatically acquired, which explains why

summary writing skills are late developing.

Although summary writing skills are very important but late in developing, it is

generally believed that summarization can be taught (Friend, 2001; Bean & Steenwyk,

1984; Kern et al., 2003). Strode (1989) reports instruction of summary writing to

students of different age groups, and Brown & Day (1983) have observed that more

mature students outperform the immature ones in summary writing instruction.

In fact, direct instruction of summary writing has been found to benefit students

in many ways (Enos, 1988; Hill, 1991; Tsai, 1995). Furthermore, according to

Casazza’s (1993) finding, students become more active in learning through direct

instruction of summary writing.

Radmacher and Latosi-Swain (1995) come to the conclusion that students can

comprehend a text at a higher level if they have learned how to summarize it. In

addition, Smith & Dahl (1984) and Day (1986) point out that students can benefit

most from detailed and explicit training; this is particularly true for those who are of

low proficiency in academic performance.

2.4 The Effects of Summary Writing Instruction

Direct instruction of summary writing is without doubt of great importance in

language teaching and learning, and many scholars and researchers have repeatedly

pointed out its beneficial effects. For example, Strode (1989) observed that students

perform better in their academic fields in college if they are taught how to summarize

a text.

Other researchers have found that students show better comprehension of the

text they are required to read if they have developed sufficient summary writing

ability (Friend, 2001 ; Enos, 1988; Casazza,1993; Radmacher & Latosi-Sawin, 1995).

Adrapoff (1970) also cites in his study that the better understanding students have of a

passage, the better they can summarize it. In addition, Davis & Hult (1997) have

found that students display better retention of the material they read after they are

required to summarize the text.

Many researchers have also observed that teaching students how to summarize

in class can help to improve their comprehension and recall (Brown, Day & Jones,

1983; Crews, 1983; Strode, 1989; Fitzgerald& McBeth, 1991; Swanson & De La Paz,

1998). What’s more, Hill (1991) points out that mastering summary skills can help

students to remember the material, strengthen comprehension and reinforce their

vocabulary.

Duke & Pearson (1985) and Amuchie (1983) also state that, after effective

instruction of summary writing, not only students’ comprehension ability would

become better, their summarization ability could be improved as well. Still other

researchers hold the view that learning how to summarize a text provides students

opportunities for them to learn to organize information (Smith & Dahl, 1984), force

them to be more attentive (Barnet et al., 1988), and engage them in learning on

cognitive levels (Sternglass, 1983).

It is worth noting, however, that as Bermudez & Prater (1998) point out in the

abstract page, “merely incorporating the summary-writing exercise did not improve

comprehension or retention, but those students who could identify main ideas and

incorporate them into the summaries perform better on the comprehension and

retention measures.”

Based on the findings of the above-mentioned researches, it is clear that most of

the researches are conducted from the perspective of investigating the possible effects

of English summary writing instruction on enhancing the participants’ reading

comprehension or even retention abilities. Very few of the researches explore its

possible effects on improving the participants’ writing abilities or even summative

abilities. Therefore, in the study, we want to probe into the effects of English

summary instruction on the senior high school students’ reading comprehension

abilities, general writing abilities and their summary writing abilities as well.

2.5 The Procedure of Summary Writing Instruction

There are concrete steps to follow in summary writing instruction. Friend

(2002), for example, proposes that teaching summary writing involves three steps.

The first step is getting the students to read the whole text as a whole. That is, before

asking them to write a summary, the teacher should have the students read the text

carefully to make sure that they really understand it. Then, the teacher should ask the

students to locate the thesis of the text. The last step involves getting the students to

determine which are the major supporting ideas for the thesis.

However, from the pedagogical point of view, summary writing instruction

should be implemented as follows. To begin with, teachers should explain what is

meant by terms such as “important information,” “trivial or redundant information,

and topic sentences, etc” (Casazza, 1993; Swanson & De La Paz, 1998). Second,

teachers should present to the students the basic steps or rules involved in

summarizing a given passage (Casazza, 1993). Last, students should be given ample

opportunities to apply the rules they have learned so as to familiarize themselves with

the basic procedures in writing a summary (Casazza, 1993; Swanson & De La Paz,

1998). Furthermore, as Swanson & De La Paz (1998) correctly point out, students

need to be constantly given guidance and feedback form the teacher as they attempt to

create their own summaries.

In this study, we will adopt the procedure of summary writing instruction

proposed by Casazza (1993) and Swanson & De La Paz (1998). The instructor will

explain what the six rules (see below) mean first, and then demonstrate to the subjects

how to put the rules to use in summarizing a text. Afterwards, the subjects will be

given sufficient opportunities to practice the rules they have learned. The instructor

will also offer constant guidance to those who have difficulty in the process of the

treatment.

2.6 Two Major Approaches to Summary Writing Instruction

As Casazza (1993) points out, there are generally two major approaches

nowadays regarding the direct instruction of summary writing. One approach is

commonly referred to as GIST (Generating Interactions between Schemata and Text),

proposed by Cunningham in 1982. The procedure in GIST, as outlined in Enos (1988),

is as follows:

The GIST procedure restricts the length of summaries to fifteen blanks.

Students begin at the sentence level, writing new fifteen word sentences as more information is revealed, deleting trivial information and redundancies until a fifteen-word summary is produced for the entire paragraph. The process continues through an entire passage. (p.12)

Swanson & De La Paz (1998) view GIST as a form of “gist summary”, and

many researchers (for example, Duke & Pearson, 1985; Strode, 1989; Enos, 1988; and

Bean & Steenwyk, 1984) have employed this approach in their studies. Bean and

Steenwyk (1984), in particular, point out the effectiveness of GIST in increasing

students’ reading comprehension.

The other approach is generally referred to as the Rule Governed Approach, and

contains six macro-rules, as described in Bean and Steenwyk (1984):

The rule-governed approach to teaching summarization described by

McNeil and Donant (1982) introduces students to the six macrorules for comprehension advanced by Kintsch and van Dijk (1978). These are (a) deleting unnecessary or trivial material, (b) deleting material that is important but redundant, (c) substituting a superordinate term for a list of items, (d) substituting a superordinate term for components of an action, (e) selecting a topic sentence, and (f) when there is no topic sentence, inventing one. (p.298)

These six rules are further merged into three in Brown & Day (1983) and

Friend (2001) while in Casazza (1993) they are categorized into four types. There are

two points in common in these two new frameworks. First, they both point to the fact

that two of the original six rules involve deletion of minor and redundant details.

Second, they also point to the fact that two of the original rules involve combining

similar details into categories and providing a superordinate term. In Casazza (1993),

selection of main idea sentences is separated from invention of the sentences, but in

Brown & Day (1983), these two rules are thought of as having to do with “providing a

summary of a main constituent unit of text, the paragraph” (p. 2).

The Rule Governed Approach has been a quite popular framework of research

and many scholars have conducted their research employing this approach, such as

Strode (1989), Enos (1988), Day (1986), Swanson & De La Paz (1998), Duke &

Pearson (1985), and Amuchie (1983). Moreover, as Duke & Pearson (1985) state

convincingly, “through teacher modeling, group and individual practice, students

learn to apply these rules to create brief summaries of text” (p.103). Furthermore,

Approach is more beneficial than GIST for the participating students.

Because the subjects in this study are not highly proficient in learning English

as a foreign language, we believe that it is impossible for the subjects to summarize a

text well by merely giving them a paragraph and then asking them to summarize the

text into a fifteen-word sentence as specified in GIST. Besides, we believe that with

explicit instruction of the six rules of the Rule Governed Approach, the subjects can

have better idea of how to summarize a text even on occasions where the instructor’s

help is not available. Therefore, in the study we adopt the Rule Governed Approach

when we give instruction in class to teach the experimental subjects how to

summarize a text.

2.7 Two Major Frameworks for Assessing a Summary

Different frameworks for assessing the effectiveness of a summary have been

proposed. Below we present two frameworks that are commonly employed by many

researchers that adopt the Rule Governed Approach in their studies. The first

framework, proposed by Garner (1982), is designed to assess the summary abilities of

undergraduate college students. In his study, Garner separated the participants into

two groups: high-proficient or low-proficient, according to their performance in the

summary writing task. This assessing framework focuses on the number of important

ideas and the number of words used when assessing a summary. The important ideas

are decided by having graduate students rate the sentences in the text as important,

moderate or unimportant.

In the other framework, recommended by Friend (2001), a summary is assessed

based on the following criteria: “inclusion of predetermined important ideas,

exclusion of predetermined unimportant ideas, construction of a thesis statement,

sentence transformation, and stating the full main idea” (p.327).

In the present study, we adopt the assessment framework recommended by

Friend (2001) when assessing the subjects’ summaries. We will take into

consideration whether predetermined important ideas are included and whether there

is a topic sentence, as well as whether the subjects state explicitly the main ideas of

the passage they read.

2.8 Summary

In the first part of this chapter, we reviewed previous studies on the major

features and classification schemes of summaries. As Enos (1988) points out that

summary writing is a very active process, the importance of summary writing

instruction can not be overemphasized too much. But to the senior high school

students participating in the present study, this is their first time to learn how to

summarize; therefore, the genre of summary chosen for them to write in this study

belong to the so-called writer-based summaries as defined by Hill (1991).

We also briefly mentioned the necessity and the effect of summary writing

instruction. Smith & Dahl (1984) and Day (1986) propose that through explicit and

detailed instruction of summary writing, students of low proficiency level can benefit

a lot. In addition, Duke & Pearson (1985) and Amuchie (1983) also state that, after

effective instruction of summary writing, not only students’ comprehension ability

would become better, their summarization ability could be improved as well.

We further introduced the procedures and two major approaches of summary

writing instruction—GIST and Rule Governed Approach. In the present study, we will

implement the instructing procedures from explanation to modeling to application

(Casazza, 1993; Swanson & De La Paz, 1998). Besides, we will adopt the Rule

Governed Approach proposed by Bean and Steenwyk (1984). Finally in this chapter,

we presented two major frameworks for assessing a summary and Friend’s framework

(2001) will be used in the study.

It must be pointed out that research on the techniques for teaching summary

writing to senior high school students is in fact quite scarce. Most of the exiting

studies lay much emphasis on the effect of summary writing instruction on improving

the participants’ reading comprehension abilities; seldom have the researchers applied

either GIST or Rule Governed Approach to investigating the effect on polishing the

participants’ summary writing abilities. It is apparent that this is a fairly new field and

more research is urgently needed.