J Behav Educ (2012) 21:80–98

DOI 10.1007/s10864-011-9137-6

ORIGINAL PAPER

How and to What Extent Do Two Cover, Copy,

and Compare Spelling Interventions Contribute
to Spelling, Word Recognition, and Vocabulary
Development?

Kathryn E. Jaspers • Robert L. Williams •

Christopher H. Skinner • David Cihak •

R. Steve McCallum • Dennis J. Ciancio

Published online: 27 September 2011

 Springer Science+Business Media, LLC 2011

Abstract We used an adapted alternating treatments design to evaluate and

compare the effects of 2 spelling interventions on spelling acquisition and main-
tenance, word reading, and vocabulary in three first-grade students. The first
intervention, Cover, Copy, and Compare (CCC), involved having participants look
at a word, cover it, write it, then compare the written response with the original
stimulus. For the second intervention, Cover, Copy, and Compare ? Sentence
Definition (CCC ? SD), CCC was supplemented with the experimenter reading a
sentence containing the word and a brief definition of the word. Results showed that
both interventions increased participants’ spelling at an equivalent rate, which was
greater than a control condition. All participants showed greater gains in word
reading in the 2 interventions than the control condition, and only 1 participant was
better able to define words learned in the CCC ? SD condition relative to the CCC
condition.

Keywords Spelling
Cover copy compare
Generalization
Maintenance

Efficiency

Introduction

Cover, Copy, and Compare (CCC) is a research-based strategy that has been shown
to enhance learning across skill and content areas (e.g., spelling, mathematics,
science, and geography) and students, including elementary and secondary students

K. E. Jaspers (&)
School of Human Sciences and Humanities, University of Houston, Clear Lake,
2700 Bay Area Blvd, Suite B-1508, Houston, TX 77058-1098, USA
e-mail: jaspers@uhcl.edu

R. L. Williams
C. H. Skinner
D. Cihak
R. S. McCallum
D. J. Ciancio

University of Tennessee, Knoxville, TN, USA

123
J Behav Educ (2012) 21:80–98 81

and students with intellectual disabilities, learning disabilities, and emotional and
behavioral disorders (Grskovic and Belfiore 1996; Hubbert et al. 2000; McLaughlin
and Skinner 1996; Murphy et al. 1990; Nies and Belfiore 2006; Skinner et al. 1989,
1992). Skinner et al. (1997b) described the CCC process as looking at the stimulus,
covering the stimulus, copying the stimulus, and comparing the written copy to the
original stimulus. If the student determines that the response and stimulus match,
then he/she progresses to the next item. If the response does not match (error), the
student writes the correct response a set number of times. Correction procedures
may involve copying the word following an error only once or copying the word
multiple times (i.e., positive practice overcorrection).
A number of components of the CCC intervention have been shown to be
effective in increasing performance in spelling, including immediate self-evaluative
feedback and daily practice. The CCC intervention provides students with
immediate corrective feedback on performance, which has been found to be more
effective than traditional spelling approaches that generally involve practice without
checking whether the practice is accurate (e.g., Grskovic and Belfiore 1996; Hubbert
et al. 2000; Nies and Belfiore 2006). In one study, Nies and Belfiore (2006)
compared the CCC strategy to a copy-only strategy. In the copy-only condition,
students were asked to say the word, point to the word, repeat the word, and copy
the word. This method differed from CCC in that it did not require the student to
cover the word while writing it and compare the original stimulus with his or her
own response. Nies and Belfiore found that CCC was more effective than the
traditional copy-only spelling strategy in increasing spelling performance (words
learned and words retained). When working with students who have spelling skills
deficits, the self-evaluation process may enhance the effectiveness of immediate
feedback (Okyere et al. 1997), especially when this self-evaluation results in more
informative feedback (e.g., provides correct spelling or precisely indicates errors),
as opposed to merely providing accuracy information (Bosman et al. 2006).
Another important aspect of CCC interventions for spelling is that educators can
use flow word lists that allow students to learn at their own pace. McLaughlin et al.
(1991) compared fixed lists of spelling words (traditional pretest and posttest) to
flow word lists with daily CCC practice. Flow word lists can be tested daily, and
once a child has spelled a word correctly a set number of times (typically 2 or 3
consecutive days), the word is replaced by a new word (see McLaughlin et al. for a
description of the Add-A-Word spelling program). This approach has been shown to
more effectively improve students’ spelling than a traditional weekly approach
(Murphy et al. 1990; McLaughlin et al. 1991).

Generalization and Efficiency

There is an expanding body of research demonstrating the relationship between

spelling and other related skills, including reading (Noell et al. 2006; de Rose et al.
1996; Graham et al. 2002) and writing (Berninger et al. 2002; Graham et al. 2002).
Orthographic representation of a word in memory will enhance the speed and
accuracy with which it is recognized (Ehri 1998; Perfetti 1992). As such, the writing
of words supports the reading of words. For example, Uhry and Shepherd (1993)

123
82 J Behav Educ (2012) 21:80–98

found that students demonstrated greater gains in spelling, segmenting, word

reading, nonsense word reading, and passage reading when instruction focused on
spelling and segmenting of words rather than simply reading letters, words, and text.
Correct spelling of a word is predictive of concurrent word reading (Foorman and
Francis 1994) and also shows utility screening for later reading difficulty (Foorman
et al. 1991; Foorman and Ciancio 2005).
Vocabulary has been identified as one of the five components of effective reading
instruction, with acquisition of vocabulary occurring both directly and indirectly
(National Institute of Child Health, Human Development 2000). When taught
directly, vocabulary instruction should involve definitional and contextual infor-
mation, repeated exposure to meaningful information about the vocabulary word,
and active child participation (Sedita 2005). Two of these methods that can be easily
added to a spelling intervention include providing a definition and providing context
(by using the word in the sentence). A combination of definitional and contextual
information may provide greater increases in vocabulary development than either
method used alone (National Institute of Child Health, Human Development 2000).
Providing definitions and sentences that contain target words may enhance
acquisition, maintenance, and generalization (e.g., enhance vocabulary, compre-
hension, and readings skills); however, such procedures may also increase
instructional time.
Once a skill has been learned, subsequent goals include maintenance over time
and generalization of accurate responding (Haring and Eaton 1978). However, few
researchers have assessed for generalization and maintenance while also consid-
ering intervention efficiency (Bramlett et al. 2010; Skinner and Daly 2010). One
exception was provided by Nist and Joseph (2008), who examined the effects of
flash card reading strategies on word reading, maintenance, and students’ ability to
read words in sentences (near generalization). Two interventions may be equally
effective when keeping number of trials constant, but the intervention that requires
less time can result in significantly faster rates of skill development (see Cates et al.
2003; Nist and Joseph 2008; Skinner et al. 1995; Skinner 2008).

Purpose

Although researchers have collective evidence that suggests a number of procedures

are effective in improving students’ spelling (e.g., Gordon et al. 1993) and that a
relationship exists between spelling and other literacy variables (Graham et al.
2002; Noell et al. 2006), few researchers have examined the differential
effectiveness of spelling interventions on other related variables. In the current
study, we compared the effect of two spelling interventions. One intervention was
CCC; the other was CCC supplemented with the experimenter reading a short
definition of the word and using it in a sentence (CCC ? SD). Although using the
word in a sentence and/or giving a definition has often been embedded in spelling
instruction and/or assessment (e.g., Grskovic and Belfiore 1996; Murphy et al. 1990;
McGuffin et al. 1997), researchers have not evaluated the effect of providing
definitions or words in sentences.

123
J Behav Educ (2012) 21:80–98 83

Working with three first-grade students, we used an adapted alternating

treatments design (Sindelar et al. 1985) to determine if supplementing CCC with
definitions and sentences (CCC ? SD) resulted in a more efficient procedure for
enhancing spelling performance (accuracy and maintenance). Additionally, we
conducted generalization assessments by measuring the impact of these procedures
on reading performance and vocabulary development (Nist and Joseph 2008;
Skinner and Daly 2010).

Method

Participants and Setting

Participants were three first-grade African American male students enrolled in an

urban after-school program. Approximately 30 children from the local elementary
school in Kindergarten through fifth grade participated in the after-school program
each day. Parents could enroll their children in the after-school program, often at the
recommendation of a child’s teacher or case worker with the local Department of
Children and Families. The after-school program cost $61 per week, but more than
90% of the children paid less than the full amount due to need-based financial
assistance. The program was designed to provide a safe and structured after-school
setting for children who were from a low-income area. During the program, children
worked on homework, participated in group athletic games, played with toys or
board games, or read under the supervision of employees and volunteers.
Participants for this study were recruited by first discussing the spelling
intervention with the after-school program director. The program director provided
the names of four first-grade students who attended the after-school program
regularly. The program director provided a consent form to the parent of each child.
Consent forms were returned for four students, though one student stopped
attending the after-school program after consent was received but before beginning
the intervention; thus, three students participated in this study. The experiment was
conducted at a table in the office of the after-school program director, next to the
room in which the children participated in the after-school program.
Each participant chose a pseudonym for himself, which was used on all
experimental materials. The three names chosen were Kirk, Manny, and Mike. Each
child was 6 years old at the onset of the study. The study occurred late in the school
year, so each participant had nearly completed first grade. Because this was a
YMCA-run after-school program, teachers could not be contacted in order to access
participants’ grades and classroom performance; therefore, the spelling and word-
reading sections of the Woodcock-Johnson III Tests of Achievement (WJ-III;
Woodcock et al. 2001) were administered as a pretest in order to measure each
participant’s normative score in spelling and word reading. Kirk scored 100 on
the Spelling subtest and 92 on the Word-Reading subtest. Manny scored 97 on the
Spelling subtest and 105 on the Word-Reading subtest. Mike scored 96 on the
Spelling subtest and 97 on the Word-Reading subtest. Thus, each participant’s
performance on both subtests was in the average range. Participants’ average scores

123
84 J Behav Educ (2012) 21:80–98

on these measures suggest basic academic ability and prerequisite skills conducive
to engaging in the spelling intervention, as state and national standards for the end
of first grade include fluently reading connected text, decoding phonetically regular
unknown words, having a developing sight word vocabulary, and correctly spelling
common words.

Materials

Materials used in this study included a stopwatch, pencils, CCC worksheets, index
cards (used to cover the target word), pretest and posttest measures, word lists, and
assessment probes. The CCC worksheet was an 8 9 11 in. sheet of paper with five
columns. The first column contained the spelling word, and the second column
provided a space for the participant to write the word. The final three columns were
used for the participants to re-write the word following an initial error. Word lists
were comprised of unknown words from Graham, Harris, and Loynachan (1993),
which consists of grade-level word lists often found in children’s reading, spelling,
and writing. Graham et al. identified 94 first-grade words and 296 second-grade
words. The methods used to identify each participant’s unknown words from
Graham et al.’s first- and second-grade word lists are described in the procedures
section.

Experimental Design, Variables, and Procedures

An adapted alternating treatments design (Sindelar et al. 1985) was used to evaluate
spelling performance under three different conditions (CCC, CCC ? SD, and
control). Both treatments were applied daily for 6 min each. While the phase length
was held constant, trials per session varied. To assess for learning outside of the
experimental conditions, testing effects, and spillover effects, control words were
assessed each day (Sindelar et al. 1985; Skinner and Shapiro 1989). Across all three
lists, as words were mastered (i.e., spelled correctly across two consecutive
assessments), new words were added to that condition. Previous researchers using
flow word lists have used mastery criteria of either two or three consecutive
assessments correct (McLaughlin et al. 1991). Because the assessment in the current
study occurred at the beginning of the daily session, the participant must have
spelled the word correctly approximately 23.5 h after the most recent intervention
on two consecutive sessions for the word to be considered mastered.

Daily Spelling Data

The primary dependent variable was the cumulative number of spelling words
mastered. Each day, the session began with the participants being assessed on 18
words, the 12 words targeted the previous day during the two treatments (six words
in each condition) and the six control words. When a word was mastered, it was
replaced with the next unknown word from the participant’s word list. These data
were graphed in time-series format, and visual analysis was used to interpret the
results (Kazdin 2011).

123
J Behav Educ (2012) 21:80–98 85

Daily Intervention Performance Data

Following the daily assessment, participants spent 6 min engaged in CCC and 6 min
engaged in CCC ? SD. During intervention sessions, data were collected on (a) the
total number of trials in each session, (b) the number of spelling errors in each
session, and (c) the number of words spelled correctly in each session (trials correct
the first time practiced). As previous researchers have shown opportunities to
respond influence learning (see Skinner et al. 1996 for a review), these data were
collected to determine whether adding sentences and definitions reduced learning
trials and rates of accurate responding.

Pretest/Posttest, Maintenance, and Generalization Data

Pretest and posttest data were collected in the three sessions immediately prior to
beginning treatment and two sessions immediately after the final treatment.
Approximately 2 and 4 weeks after the final treatment, identical procedures were
used to collect maintenance data. Additional pretest/posttest and maintenance
generalization data were collected. Specifically, participants were provided with the
list of words and asked to read each word aloud and provide a definition of the word.
Both total number of words correct and percent correct were calculated for each
condition across the pretest, posttest, and two maintenance assessments.

General Procedure

Each day, the primary experimenter worked one-to-one with each participant. Over
the first three sessions, pretesting procedures were used to select 60 words for each
participant and then assess students’ ability to read and define these words. As a
result of each participant’s errors in spelling, word reading and definitions, each
participant’s word list was unique to the individual. Over the next 13 days,
experimental procedures were run. With the exception of the first day, sessions
began by assessing participants’ spelling performance on the previous session’s 18
words (six per condition) assigned to the three conditions. Next, the two treatment
procedures were applied for 6 min each (see the ‘‘Appendix’’ for a procedural
integrity checklist which outlines the daily order for tasks). After the final
experimental session, participants were asked to spell, read, and define all words
from their word lists. Maintenance and generalization data on spelling, word
reading, and definitions were assessed again 2 and 4 weeks following the
experiment. The following sections describe the pretest, experimental phase, and
posttest procedures in greater detail.

Pretest Procedures

The pretest and all other assessments were administered to each participant
individually. The experimenter first administered the spelling and reading sections
of the WJ-III in order to evaluate participants’ normative spelling and reading
achievement. Participants were then given a spelling pretest over two sessions using

123
86 J Behav Educ (2012) 21:80–98

Graham et al.’s (1993) first-grade and second-grade words. Words that could not be
easily defined (e.g., ‘‘of’’) or had multiple spellings (e.g., ‘‘be’’ vs. ‘‘bee’’) were
excluded. For the spelling pretest, the experimenter read a word, the participant
wrote down the word, and the experimenter read the next word after the student
finished. The experimenter tracked the participants’ incorrect responses, and once
each participant reached 60 unknown words, testing was stopped. First-grade words
were assessed before progressing to second-grade words, and word order for the
pretest was randomized within grade-level lists.
On the third pretest day, words were presented to the participants typed on
8 9 11 in. paper with 20 words per page, and participants were asked to read each
of the 60 words from their word list, which established pretest baseline levels of
word recognition. Participants’ word reading was not timed, and they were
instructed to do their best reading. After the participants finished reading their word
lists, participants were asked to provide an oral definition of each word, which was
audiotaped for the purpose of scoring. For each word, the experimenter read the
word and asked the participant what the word meant. Words were scored as correct
if the participant provided an accurate definition, synonym, antonym (e.g., ‘‘stop
means don’t go’’), or category (e.g., ‘‘apple is a fruit’’).
Using stratified random assignment, the unknown spelling words were then
divided into three equivalent lists of spelling words. Words were stratified across
(a) the word’s grade level (first or second grade), (b) the number of letters in the
word (range: two to seven), and (c) whether the participant was able to read and/or
define the word. Word lists were then randomly assigned to each of the three
conditions.

Experimental Phase

Three conditions were used: CCC, CCC ? SD, and control condition. Each
condition began with six unknown target words. In the CCC condition, participants
were given a CCC worksheet with the six target words listed. The experimenter read
the first word to the participant three times, then had the participant look at the
word, cover the word with an index card, write the word in the next column, and
then compare his spelling to the correct spelling. If the participant spelled the word
correctly, the experimenter and participant continued to the next word. If the
participant incorrectly spelled the target word, the participant then wrote the correct
spelling three times in the designated space on the CCC worksheet as an
overcorrection technique. In the CCC ? SD condition, participants were given a
CCC sheet identical to that used in the previous condition. Procedures in this
condition were the same, except that after the experimenter said the word aloud the
first time, the experimenter then used the word in a sentence, repeated the word, and
gave a brief definition. Thus, across both conditions, the experimenter stated the
word three times. The words assigned to the control condition were assessed to
measure testing and/or history effects (Skinner and Shapiro 1989).
Each day began with an assessment of the 12 words from the previous day’s
worksheets (six from the CCC worksheet and six from the CCC ? SD worksheet)
and the six words assigned to the control condition, with word order randomized.

123
J Behav Educ (2012) 21:80–98 87

Once the participant correctly spelled a word on two consecutive assessments, that
word was dropped and a new word was randomly selected from the participant’s
unknown word list. After the assessment, the experimenter and participant
completed the two interventions, with each intervention occurring for 6 min. The
participant repeatedly practiced the same six words for the entire 6 min of the first
intervention, and then practiced the second set of six words for the entire 6 min of
the second intervention. The entire intervention time (including both conditions and
an assessment of the previous day’s words and the control list) lasted approximately
20 min. To control for sequencing effects, the two treatments were presented in a
counterbalanced order across days. Because each child had his own list of unknown
words, sessions were conducted individually.

Posttest and Maintenance Procedures

Posttest data were collected 2 days after the final treatments, and maintenance data
were collected 2 and 4 weeks following the final treatment session. Each of these
three assessments lasted 2 days. On the first day, the participant was asked to spell
each word targeted throughout the study across the three word lists (CCC,
CCC ? SD, and control list). Using the same list, on the second day, the
participants read a list of the words and gave an oral definition of the words.

Interrater Reliability and Procedural Integrity

The primary experimenter collected, scored, and recorded all data. A second
experimenter independently scored 20% of the CCC sheets for each condition and
20% of the daily tests, and interscorer agreement for both the CCC sheets and daily
tests was 100%. The experimenter audiotaped the participants’ definitions during
the pretest and posttests. The experimenter then typed the participants’ oral
definitions, and the second experimenter independently scored 50% of the
definitions. Interscorer agreement for participants’ definitions was 84 out of 90
words, or 93%. The second experimenter attended 20% of experimental sessions
and used a checklist (see the ‘‘Appendix’’), which listed the steps to be followed
each day, to record procedural integrity. The second experimenter checked each
item on the procedural integrity checklist, in order, as the primary experimenter
carried out the intervention. Procedural integrity was 98%.

Results

Cumulative Words Learned

Figure 1 displays the cumulative number of words mastered (i.e., spelled correctly
over two consecutive assessments) across conditions for each participant. Visual
analysis of Fig. 1 showed that across all students, CCC and the CCC ? SD spelling
interventions resulted in similar and steady increases in words mastered. Kirk
mastered 15 words in both the CCC and CCC ? SD conditions, Manny mastered 14

123
88 J Behav Educ (2012) 21:80–98

Pretest Alternating Treatments Follow-up

20 I 2 4
18
16
Kirk
Spelled Correctly
Cumulative Words

14 CCC
12
10
8
6 CCC+SD
4 Test
2
0

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
6-min Sessions
Pretest Alternating Treatments Follow-up
20 I 2 4
18
16 Manny
Cumulative Words
Spelled Correctly

14
12
10
8
6
4
2
0

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
6-min Sessions
Pretest Alternating Treatments Follow-up
20 I 2 4
18
16
Cumulative Words

Mike
Spelled Correctly

14
12
10
8
6
4
2
0

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
6-min Sessions

Fig. 1 Cumulative words spelled correctly during intervention and follow-up. I = immediate posttest,
2 = 2-week follow-up, 4 = 4-week follow-up

words in both the CCC and CCC ? SD conditions, and Mike mastered 18 words in
the CCC condition and 19 words in the CCC ? SD condition. For all three
participants, the control condition remained at near baseline level, which suggests
that spillover effects were minimal and that the interventions as opposed to some
other variable(s) caused the increases in spelling accuracy on CCC and CCC ? SD
words (Sindelar et al. 1985; Skinner and Shapiro 1989). As Fig. 1 shows no

123
J Behav Educ (2012) 21:80–98 89

consistent divergence across the two interventions for any participant, these results
suggest that CCC and CCC ? SD interventions were equally effective at enhancing
spelling acquisition rates.

Intervention Performance Data

The totals, means, and standard deviations for (a) trials per 6-min intervention session,
(b) errors per intervention session, and (c) trials correct on the first attempt per session
for the two treatments are displayed in Table 1. Graphs of the number of trials and
errors per session per condition are found in Fig. 2. Kirk engaged in just over three
trials per session more in the CCC condition than in the CCC ? SD condition. Manny
engaged in 0.87 trials per session more in the CCC condition than in the CCC ? SD
condition, and Mike engaged in 1.3 trials per session more in the CCC ? SD than in
the CCC condition. Visual analyses of time-series graphs indicate significant overlap
in the CCC and CCC ? SD conditions for Manny and Mike. The three participants
ranged from 0 to 3 errors per session in each condition. Kirk and Mike made an
approximately equivalent number of errors in each condition. Manny committed 0.46
more errors per day in the CCC ? SD condition than in the CCC condition. Because
errors could influence the number of trials (due to the overcorrection procedure), the
number of trials per session that were correct the first attempt was computed. Overall
error rates were low and consistent across conditions; thus, review of trials correct in
the first attempt indicated similar patterns to trials correct.

Pretest/Posttest and Maintenance: Spelling, Word Reading, Definitions

Participants’ pretest performance was compared with their performance on an

immediate posttest, 2 weeks’ follow-up, and 4 weeks’ follow-up for the following

Table 1 Trials, errors, and trials correct first attempt per session
Kirk Manny Mike

CCC CCC ? SD CCC CCC ? SD CCC CCC ? SD

Trials
Total 217 175 151 140 229 246
Mean 16.69 13.46 11.62 10.77 17.62 18.92
SD 5.72 5.61 3.88 3.81 4.79 3.55
Errors
Total 17 16 10 16 16 11
Mean 1.31 1.23 0.77 1.23 1.23 0.85
SD 1.11 0.73 1.01 0.93 1.24 0.80
Trials correct first attempt
Total 200 159 141 124 213 235
Mean 15.38 12.23 10.85 9.54 16.38 18.08
SD 6.61 6.17 4.71 4.43 5.65 4.05

123
90 J Behav Educ (2012) 21:80–98

30
CCC+SD trials
25 Kirk CCC trials
Trials and Errors

20
15
10 CCC errors
CCC+SD errors
5
0
1 2 3 4 5 6 7 8 9 10 11 12 13

6-min Sessions
30
Manny
25
Trials and Errors

20
15
10
5
0
1 2 3 4 5 6 7 8 9 10 11 12 13

30 6-min Sessions

25 Mike
Trials and Errors

20
15
10
5
0
1 2 3 4 5 6 7 8 9 10 11 12 13
6-min Sessions

Fig. 2 Trials and errors per session

variables: spelling, word reading, and definitions. Table 2 includes the percentages
correct for spelling, word reading, and definitions. See Table 3 for the total number
correct for spelling, word reading, and definitions.

Spelling

Spelling improvements were measured by computing percentage and total number

of words learned in each condition. Letter sequences correct were also computed but

123
J Behav Educ (2012) 21:80–98 91

Table 2 Percentage correct for spelling, word reading, and definitions

Kirk Manny Mike

CCC CCC ? SD Control CCC CCC ? SD Control CCC CCC ? SD Control

Spelling
Pretest 0 0 0 0 0 0 0 0 0
Immediate 62 43 22 40 50 20 68 81 14
2 weeks 24 38 26 40 50 8 48 48 24
4 weeks 38 33 22 45 44 20 44 53 33
Word reading
Pretest 19 19 30 55 61 64 32 38 48
Immediate 48 43 22 80 78 60 88 95 76
2 weeks 57 48 39 85 83 64 92 95 71
4 weeks 43 57 43 80 94 68 96 90 81
Definition
Pretest 19 19 30 55 56 52 36 33 29
Immediate 28 62 22 35 67 48 40 38 43
2 weeks 33 67 26 40 56 56 32 38 43
4 weeks 28 57 22 35 50 60 36 33 33

Table 3 Total number correct for spelling, word reading, and definitions
Kirk Manny Mike

CCC CCC ? SD Control CCC CCC ? SD Control CCC CCC ? SD Control

Spelling
Pretest 0 0 0 0 0 0 0 0 0
Immediate 13 9 5 8 9 5 17 17 3
2 weeks 5 8 6 8 9 2 12 10 5
4 weeks 8 7 5 9 8 5 11 11 7
Word reading
Pretest 4 4 7 11 11 16 8 8 10
Immediate 10 9 5 16 14 15 22 20 16
2 weeks 12 10 9 17 15 16 23 20 15
4 weeks 9 12 10 16 17 17 24 19 17
Definition
Pretest 4 4 7 11 10 13 9 7 6
Immediate 6 13 5 7 12 12 10 8 9
2 weeks 7 14 6 8 10 14 8 8 9
4 weeks 6 12 5 7 9 15 9 7 7

mirrored percentage and total words learned and thus did not provide unique
information. Overall, each participant showed increased levels of spelling on the
experimental lists immediately after the intervention and at the subsequent
maintenance probes. Each participant also improved on the control lists, though

123
92 J Behav Educ (2012) 21:80–98

at much lower percentages. Manny correctly spelled a relatively low percentage of

the words learned immediately following the intervention (40% of CCC and 50% of
CCC ? SD words) but maintained that level over the 2 and 4 weeks’ posttests.
Conversely, Mike demonstrated a high level of spelling recall immediately
following the intervention (68% in the CCC condition and 81% in the CCC ? SD
condition) but regressed by the 4-week posttest (44% in CCC and 53% in
CCC ? SD). Immediately after the intervention, Kirk was better able to spell words
in the CCC condition (62% of words) than in the CCC ? SD condition (43% of
words), but this difference disappeared by the 4-week posttest.

Word Reading

See Table 2 for percentages of words read correctly and Table 3 for total number
of words read correctly. All three participants showed gains in the percentage of
words read correctly in the CCC (24% gain from pretest to 4-week posttest for
Kirk, 25% for Manny, and 64% for Mike) and CCC ? SD conditions (38% for
Kirk, 33% for Manny, and 52% for Mike). Also, all three demonstrated increases
in words read correctly from the control list (13% for Kirk, 4% for Manny, and
33% for Mike), although these rates were less than those found in the CCC and
CCC ? SD lists.

Definitions

Participants’ results for percentage and total number of words defined correctly can
be found in Tables 2 and 3. From pretest to the 4-week follow-up, Kirk correctly
defined only 9% more words in the CCC list and 8% less in the control list,
whereas he correctly defined 38% more words from the CCC ? SD. After
4 weeks, he correctly defined 12 words from the CCC ? SD list and only six
words from the CCC list. Thus, Kirk was able to define words from the CCC ? SD
condition much more effectively than from the CCC or control conditions. From
the pretest to the 4-week follow-up, Manny correctly defined 20% less in the CCC
condition, 6% less in the CCC ? SD condition, and 8% more in the control
condition. Manny’s total words defined in the CCC and CCC ? SD condition
actually declined across the course of the intervention. Similarly, Mike showed 0%
improvement from pretest to posttest in the CCC and CCC ? SD conditions and a
4% improvement in the control condition. In summary, only one of the three
participants was better able to define words learned in the CCC ? SD condition
than in the other two conditions.

Treatment Acceptability

After the last day of data collection, the experimenter asked each participant which
intervention (1) he liked better, (2) was harder, and (3) helped him learn words
better. Each participant answered that he liked the CCC ? SD intervention better
than the CCC intervention, that the CCC intervention was harder than the

123
J Behav Educ (2012) 21:80–98 93

CCC ? SD intervention, and that the CCC ? SD intervention helped him learn
words better than the CCC intervention. Thus, treatment acceptability data favored
the CCC ? SD intervention as the more preferred method.

Discussion

The current results supported the efficacy of CCC for enhancing spelling accuracy,
as all students showed steady improvement in spelling words assigned to both
interventions relative to the control words. The small increase in spelling accuracy
on untargeted words suggests that the interventions, as opposed to some other
variables (e.g., learning in class), caused the increases in spelling accuracy on
targeted word lists and also suggests that spillover effects were minimal (Sindelar
et al. 1985; Skinner and Shapiro 1989). Thus, this study provides support for
previous findings that showed the CCC add-a-word program was effective for
increasing spelling accuracy (McLaughlin et al. 1991; Murphy et al. 1990; Nies and
Belfiore 2006).
Though participants demonstrated improvements in spelling accuracy in both
conditions, review of maintenance data at 2 and 4 weeks after the interventions
suggests that some words considered mastered in the intervention were not
maintained over time. In this study, words were considered mastered when the
participant correctly spelled a word for two consecutive days on the daily
assessments. More stringent mastery criteria may increase the likelihood of long-
term maintenance.
The current results extend previous research on CCC by evaluating generaliza-
tion to word reading and vocabulary. All three participants showed greater gains on
all three word-reading posttests (immediate, 2-week follow-up, and 4-week follow-
up) on words targeted by CCC and CCC ? SD relative to control words. These
results support those of previous researchers who found that increases in spelling
accuracy can generalize to word reading accuracy (Noell et al. 2006). Comparisons
across the two interventions revealed little difference on word-reading posttest
accuracy, which suggests that including word definitions and using words in a
sentence did not enhance generalization and maintenance of word reading.
Comparisons of time-series data showed no differences in spelling acquisition
rates across the two interventions. Thus, hearing the definition of a word or hearing
it used in a sentence did not enhance spelling accuracy. However, adding these two
components to CCC did not hinder students’ spelling acquisition rates, either.
Because increasing CCC learning trials has been shown to enhance learning
(Skinner et al. 1997a), this result can be explained by examining the number of
learning trials during each intervention. Participants were expected to complete
many more learning trials in 6-min sessions under the CCC condition relative to the
CCC ? SD condition. However, learning trials per session did not differ across the
two conditions consistently for the three participants. Thus, equivalent learning rates
for the two interventions are consistent with the finding that each participant
engaged in a similar number of learning trials in the first intervention as he did in the
second intervention.

123
94 J Behav Educ (2012) 21:80–98

Limitations and Directions for Research

Future researchers should address a number of methodological limitations

associated with the current study. Pretest and posttest, as opposed to experimental
procedures, were used to evaluate changes across word reading and vocabulary.
Because only three participants were used and no repeated measures were taken
(i.e., time-series data), it is not possible to determine whether the intervention
procedures caused changes in word reading and vocabulary. Future single-subject
design researchers should assess changes in word reading and vocabulary more
frequently (e.g., see spelling data in Fig. 1) to gather data that would provide more
precise measures of learning rates over time. An alternative would be to apply a true
experimental design with a larger number of participants. Another factor that may
have influenced our findings is our method used to assess vocabulary. Each
participant was asked to provide an oral definition for each word from his word list.
Future researchers may find that a different measure of vocabulary (e.g., receptive
vocabulary) would provide a better assessment of students’ knowledge of words.
Another limitation of this study was the relatively short length of the intervention,
which was constrained due to the school year ending. Future researchers could
examine increasing intervention time (min per day or total days/weeks of
intervention) to determine the best formula for maximizing mastery and mainte-
nance over time.
Several factors limit the generalizability of these findings. Because only three
participants were used, it was not possible to compare one participant’s results to
another to form general conclusions. Also, we found a number of idiosyncratic
differences between participants that cannot be explained. One participant clearly
listened to and learned the definitions to words, while the others did not. A different
participant made greater gains in word reading than the other two participants. The
third participant maintained all his gains in spelling from the immediate posttest to
the 4-week follow-up, while the other two participants demonstrated lower levels of
spelling maintenance. These idiosyncratic findings limit our ability to draw
generalizable conclusions, but also provide clear directions for future researchers.
Because our participants were first-grade students (whose penmanship may be
more effortful than older students), writing fluency may have influenced relative
learning trial rates. It is possible that with children who write more fluently, a
greater percentage of their time would be spent listening to sentences and
definitions, relative to time spent performing CCC. If this were the case, then we
would expect students to complete many more trials in the CCC condition than in
the CCC ? SD condition, which may result in greater increases in spelling accuracy
for the CCC condition. Consequently, future researchers may want to conduct
similar time-held-constant studies with more fluent writers.
Perhaps the clearest difference between the two interventions was expected to be
found in participants’ ability to define words learned in the CCC ? SD condition.
This was anticipated because in the CCC ? SD condition, the definition and a
sentence were provided for each word; however, this expected difference was not
found. In fact, only one of the three participants, Kirk, was better able to define
words in the CCC ? SD condition than in the CCC or control conditions. The other

123
J Behav Educ (2012) 21:80–98 95

two participants showed no gains in their ability to define words from their
CCC ? SD list. A number of behavioral observations of Kirk’s performance during
the interventions and assessments may explain this difference between Kirk and the
other two participants. On multiple occasions, Kirk stated the definition aloud with
the experimenter. In addition, Kirk repeated definitions from his CCC ? SD list
verbatim on the posttests, even the 4-week follow-up. He also occasionally repeated
the sentences and definitions during the daily assessment (at which time the
experimenter did not provide the sentence and definition, nor was it requested).
Finally, Kirk engaged in fewer trials per session in the CCC ? SD condition than in
the CCC condition. This indicates that trials in the CCC ? SD condition took
slightly longer, which may have been caused by Kirk repeating the definitions.
In the current study, participants were tested daily on their spelling performance,
which may have motivated the participants to attend to this aspect of the
intervention rather than the sentences and definitions presented by the experimenter.
Additional evidence of a treatment-assessment interaction can be found in the
decline and subsequent improvement in two participants’ word reading and
definitions (which were not measured during the intervention) at the immediate and
4-week follow-up. This treatment–assessment interaction could be balanced by
providing motivation (i.e., feedback, reinforcement) to attend to the words’
definitions or by testing vocabulary accuracy more regularly (i.e., daily). Addition-
ally, researchers may want to apply some of Kirk’s procedures (e.g., having the
student repeat the definition) to determine whether these procedures enhance
acquisition and maintenance of definitions.

Summary

This study was designed to determine whether supplementing spelling CCC with
two additional components (sentences and definitions) increased spelling acquisition
and maintenance and generalization to reading and vocabulary. Because these
additional components required additional instructional time, we held time constant
across the two intervention procedures so these comparisons provided a precise
evaluation of learning rates (Skinner 2008). Results suggested that CCC enhanced
spelling and word reading acquisition and maintenance. Although only one of the
three participants appeared to learn more definitions of words assigned to the
CCC ? SD condition, adding these two components did not hinder spelling
learning rates (acquisition, maintenance) or word reading in any participants.
Furthermore, all participants indicated that the CCC ? SD intervention was more
acceptable. Consequently, these findings suggest future researchers should continue
to investigate supplementing CCC with these components as adding them did not
slow spelling or reading learning rates and may have enhanced vocabulary
development in one student.
Many researchers fail to assess how instructional techniques impact other skills
(e.g., generalizable effects) or collect precise measures of learning rates (Bramlett
et al. 2010; Codding and Poncy 2010). When conducting academic intervention
comparison studies, both of these omissions can result in researchers supporting the
least effective treatment (Skinner 2008, 2010; Skinner and Daly 2010). The current

123
96 J Behav Educ (2012) 21:80–98

study may have broader heuristic value by demonstrating how researchers might
conduct treatment comparison and component analysis studies that identify the most
useful way to apply allotted instructional time.

Appendix

Integrity Checklist - Date:

Child’s Pseudonym:
____get child
____start watch once child is seated and ready to begin
____start stopwatch
____give child 18-item pretest over the previous day’s words
____stop stopwatch, record time
____score pretest and edit lists while child has short break
____complete first intervention*
____short break
____complete second intervention*
____stop watch and record total session time

*For CCC intervention:

____start stopwatch
____read each word three times
____prompt covering, copying, comparing, if necessary
____prompt correction, if necessary
____after 6 min, stop the stopwatch, say ‘‘Stop’’, and take worksheet
*For CCC ? SD intervention:
____start stopwatch
____read each word, say sentence, say definition BEFORE child begins copying
____prompt covering, copying, comparing, if necessary
____prompt correction, if necessary
____after 6 min, stop the stopwatch, say ‘‘Stop’’, and take worksheet

References

Berninger, V. W., Vaughan, K., Abbott, R. D., Begay, K., Byrd Coleman, K., Curtin, G., … Graham, S.
(2002). Teaching spelling and composition alone and together: Implications for the simple view of
writing. Journal of Educational Psychology, 94, 291–304.
Bosman, A. M. T., van Huygevoort, M., & Verhoeven, L. (2006). Spelling feedback in an ICT-learning
environment: Issues of proficiency, training efficiency, and transfer. International Journal of
Educational Research, 45, 341–361.
Bramlett, R., Cates, G. L., Savina, E., & Lauinger, B. (2010). Assessing effectiveness and efficiency of
academic interventions in school psychology journals: 1995–2005. Psychology in the Schools, 47,
114–125.

123
J Behav Educ (2012) 21:80–98 97

Cates, G. L., Skinner, C. H., Watson, T. S., Meadows, T. J., Weaver, A., & Jackson, B. (2003).
Instructional effectiveness and instructional efficiency as considerations for data-based decision
making: An evaluation of interspersing procedures. School Psychology Review, 32, 601–616.
Codding, R. S., & Poncy, B. C. (2010). Introduction to the special issue: Toward and explicit technology
for generalizing academic behavior. Journal of Behavioral Education, 19, 1–7.
de Rose, J. C., de Souza, D. G., & Hanna, E. S. (1996). Teaching reading and spelling: Exclusion and
stimulus equivalence. Journal of Applied Behavior Analysis, 29, 451–469.
Ehri, L. C. (1998). Grapheme-phoneme knowledge is essential for learning to read words in English. In J.
M. Metsala & L. C. Ehri (Eds.), Word recognition in beginning literacy (pp. 3–40). Mahwah, NJ:
Lawrence Erlbaum Associates.
Foorman, B. R., & Ciancio, D. J. (2005). Screening for secondary intervention: Concept and context.
Journal of Learning Disabilities, 38(6), 494–499.
Foorman, B. R., & Francis, D. J. (1994). Exploring connections among reading, spelling, and phonemic
segmentation during first grade. Reading and Writing, 6, 65–91.
Foorman, B. R., Francis, D. J., Novy, D. M., & Liberman, D. (1991). How letter-sound instruction
mediates progress in first-grade reading and spelling. Journal of Educational Psychology, 83,
456–469.
Gordon, J., Vaughn, S., & Schumm, J. S. (1993). Spelling interventions: A review of literature and
implications for instruction for students with learning disabilities. Learning Disabilities Research
and Practice, 8, 175–181.
Graham, S., Harris, K. R., & Fink-Chorzempa, B. (2002). Contribution of spelling instruction to the
spelling, writing, and reading of poor spellers. Journal of Educational Psychology, 94, 669–686.
Graham, S., Harris, K. R., & Loynachan, C. (1993). The basic spelling vocabulary list. Journal of
Educational Research, 86, 263–268.
Grskovic, J. A., & Belfiore, P. J. (1996). Improving the spelling performance of students with disabilities.
Journal of Behavioral Education, 6, 343–354.
Haring, N. G., & Eaton, M. D. (1978). Systematic instructional procedures: An instructional hierarchy. In
N. G. Haring, T. C. Lovitt, M. D. Eaton, & C. L. Hansen (Eds.), The fourth R: Research in the
classroom (pp. 23–40). Columbus, OH: Merrill.
Hubbert, E. R., Weber, K. P., & McLaughlin, T. F. (2000). A comparison of cover, copy, and compare
and a traditional spelling intervention for an adolescent with a conduct disorder. Child and Family
Behavior Therapy, 22(3), 55–68.
Kazdin, A. E. (2011). Single-case research designs: Methods for clinical and applied settings (2nd ed.).
New York: Oxford University Press.
McGuffin, M. E., Martz, S. A., & Heron, T. E. (1997). The effects of self-correction versus traditional
spelling on the spelling performance and maintenance of third grade students. Journal of Behavioral
Education, 7, 463–476.
McLaughlin, T. F., Reiter, S. M., Mabee, W. S., & Byram, B. J. (1991). An analysis and replication of the
Add-A-Word spelling program with mildly handicapped middle school students. Journal of
Behavioral Education, 1, 413–426.
McLaughlin, T. F., & Skinner, C. H. (1996). Improving academic performance through self-management:
Cover, copy, and compare. Intervention in School and Clinic, 32, 113–119.
Murphy, J. F., Hern, C. L., Williams, R. L., & McLaughlin, T. F. (1990). The effects of the copy, cover,
compare approach in increasing spelling accuracy with learning disabled students. Contemporary
Educational Psychology, 15, 378–386.
National Institute of Child Health, Human Development. (2000). Report of the national reading panel.
Teaching children to read: An evidence-based assessment of the scientific research literature on
reading and its implications for reading instruction (NIH Publication No. 00–4769). Washington,
DC: U.S. Government Printing Office.
Nies, K. A., & Belfiore, P. J. (2006). Enhancing spelling performance in students with learning
disabilities. Journal of Behavioral Education, 15, 163–170.
Nist, L., & Joseph, L. M. (2008). Effectiveness and efficiency of flashcard drill instructional methods on
urban first graders’ word recognition, acquisition, maintenance, and generalization. School
Psychology Review, 37, 294–308.
Noell, G. H., Connell, J. E., & Duhon, G. J. (2006). Spontaneous response generalization during whole
word instruction: Reading to spell and spelling to read. Journal of Behavioral Education, 15,
121–130.

123
98 J Behav Educ (2012) 21:80–98

Okyere, B. A., Heron, T. E., & Goddard, Y. (1997). Effects of self-correction on the acquisition,
maintenance, and generalization of the written spelling of elementary school children. Journal of
Behavioral Education, 7, 51–69.
Perfetti, C. A. (1992). The representation problem in reading acquisition. In P. B. Gough, L. C. Ehri, & R.
Treiman (Eds.), Reading acquisition (pp. 145–174). Hillsdale, NJ: Lawrence Erlbaum.
Sedita, J. (2005). Effective vocabulary instruction. Insights on Learning Disabilities, 2, 33–45.
Sindelar, P. T., Rosenberg, M. S., & Wilson, R. J. (1985). An adapted alternating treatments design for
instructional research. Education and Treatment of Children, 8, 67–76.
Skinner, C. H. (2008). Theoretical and applied implications of precisely measuring learning rates. School
Psychology Review, 37, 309–315.
Skinner, C. H. (2010). Applied comparative effectiveness researchers must measure learning rates: A
commentary on efficiency articles. Psychology in the Schools, 47, 166–172.
Skinner, C. H., Belfiore, P. J., Mace, H. W., Williams, S., & Johns, G. A. (1997a). Altering response
topography to increase response efficiency and learning rates. School Psychology Quarterly, 12,
54–64.
Skinner, C. H., Belfiore, P. J., & Pierce, N. L. (1992). Cover, copy, and compare: Increasing geography
accuracy in students with behavior disorders. School Psychology Review, 21, 73–81.
Skinner, C. H., Belfiore, P. J., & Watson, T. S. (1995). Assessing the relative effects of interventions in
students with mild disabilities: Assessing instructional time. Assessment in Rehabilitation and
Exceptionality, 20, 207–220.
Skinner, C. H., & Daly, E. J. (2010). Improving generalization of academic skills: Commentary on the
special series. Journal of Behavioral Education, 19, 106–115.
Skinner, C. H., Fletcher, P. A., & Henington, C. (1996). Increasing learning trial rates by increasing
student response rates. School Psychology Quarterly, 11, 313–325.
Skinner, C. H., McLaughlin, T. F., & Logan, P. (1997b). Cover, copy, and compare: A self-managed
academic intervention effective across skills, students, and settings. Journal of Behavioral
Education, 7, 295–306.
Skinner, C. H., & Shapiro, E. S. (1989). A comparison of a taped-words and drill interventions on reading
fluency in adolescents with behavior disorders. Education and Treatment of Children, 12, 123–133.
Skinner, C. H., Turco, T. L., Beatty, K. L., & Rasavage, C. (1989). Cover, copy, and compare: A method
for increasing multiplication performance. School Psychology Review, 18, 412–420.
Uhry, J. K., & Shepherd, M. J. (1993). Segmentation/spelling instruction as part of a first-grade reading
program: Effects on several measures of reading. Reading Research Quarterly, 28, 218–233.
Woodcock, R. W., McGrew, K. S., & Mather, N. (2001). Woodcock Johnson III tests of achievement.
Itasca, IL: Riverside Publishing.

123