[Spelling Progress Bulletin, Summer 1983, pp14-16]

Spelling and Handwriting: Is there a Relationship?,

by Michael N. Milone, Jr, Ph.D. James A. Wilhide, and Thomas M, Wasylyk*

* Zaner-Bloser, Inc., Honesdale, PA.

The syndicated columnist Earl Wilson once told the story of the boy who brought home a note from his teacher. The note read, "Your son's handwriting is so bad we don't know if he can spell" (Askov, Otto, and Askov, 1970, p. 109).

This incident is not isolated; it probably happens often each school year. Spelling is assessed primarily through written tests; and it is reasonable to assume that handwriting problems interfere with tudents' spelling ability. Illegible handwriting makes it unlikely that a student will recognize an error and try to correct it., Moreover, poorly written words: may be unreadable to the teacher and be marked as incorrect even though the student knew how to spell the word correctly.

Despite the logic supporting the connection between handwriting and spelling, there has been little corroborative evidence of an empirical nature. The only research that seems to confirm the existence of the relationship was conducted by Strickling (1973), who compared the oral and written performance of 136 fifth graders. She found that their mean oral spelling score on the Morrison-McCall Spelling Scale was 27.6 words correct out of a possible 40, while the mean written score was 23.3. The correlation between written spelling errors and two measures of writing legibility, were .60 and .56. The difference between written and oral spelling. results and the magnitude of the correlation coefficients were considerably greater than would be expected by chance. Strickling (1973) concluded that:

The lower mean score on the written spelling test was due mainly to handwriting errors because:

a) Words missed on the written spelling test because of handwriting errors were spelled correctly on the oral test.

b) The number of words missed on the written spelling test because of handwriting errors was positively correlated with the errors made on the handwriting tests.

c) When the words missed on the written spelling test because of handwriting errors were added to the written spelling score, the total was essentially the same as the score on the oral spelling test. (p. 3717-A).

Although the results cited above are convincing, they stand alone, as no related research has been reported. The purpose of this study was to expand Strickling's work and discover whether there actually is a relationship between handwriting and spelling.

The current research involved two stages. In Part 1, the handwriting and written spelling scores of 750 students were compared; in Part 2, Strickling's method was repeated with minor variations. There were 129 students :involved in Part 2.

Method.

Participants.


Students in 40 randomly selected sixth-grade classrooms in South Carolina participated in the studies: The proportion of females to males was 1.01 to 1, and the racial balance in the sample was representative of that found in South Carolina schools: As was previously mentioned, data were obtained from a total of 879 students: 750 in Part 1 and 129 in Part 2.

Procedures.


Participating teachers were mailed an information packet that contained all the directions and materials they needed. The teachers administered a spelling test to all the students in their class; 129 randomly selected students also completed an oral spelling test based on the same words: The order of the oral and written tests was alternated so that a practice effect would not influence the outcome of the study. Students were tested for their oral and written spelling ability using the same list of words; and no more than one week elapsed between administrations of the test., For the oral spelling test, teachers recorded exactly how the students responded.

Completed tests were returned to the researchers for scoring. The number of errors for each spelling test was tallied; the written spelling tests served as the basis for the handwriting evaluation. Legibility was assessed by two trained raters using a 1-to-5 scale, with 5 representing highly legible writing and 1 corresponding to highly illegible writing. Initial rarer agreement was approximately 85%; on any sample on which there was disagreement, the raters worked together until a mutually satisfactory score was obtained. The final agreement was 100%.

Word List.


Several criteria were used to select words for the spelling test. First, the words had to be understandable to the majority of students. To ensure this, words had to be correctly defined by 50% of the fourth graders sampled in Dale and O'Rourke's research for The Living Word (1976). Words understood by this proportion of fourth-graders should be meaningful to the vast majority of sixth-graders.

Second, the words also had to be spelling demons. These demons were selected from Johnson's (1951) classic list or from those featured in popular basal spelling series.

Third, the words had to contain letters or letter combinations that were susceptible to handwriting illegibility. The findings reported by Newland (1932) served as the guidelines for selecting these words. In his study of cursive handwriting, Newland discovered that more than 60% of illegibilities could be attributed to four types of errors:
1. Failing to close letters (a like u).
2. Closing looped strokes (e like i).
3. Looping non-looped strokes (i like e).
4. Using vertical strokes rather than rounded strokes (n like u).
Twenty words that met all these criteria were used in the spelling test; they can be found in Table 3. The test was administered in a manner with which the students were familiar; the word was read by the teacher, it was used in a standard sentence, and it was repeated.

Results.

Part 1


As can be seen in Table 1, there is a strong relationship between handwriting and spelling ability: If a student's handwriting was highly legible, then it was unlikely that the student would make more than two spelling errors. If the student had illegible handwriting, then he or she probably made many spelling errors. More than 70% of the students with a handwriting score of 1 (least legible) had 5 or more spelling errors; more than 50% had 10 or more errors. The pattern was consistent across the range of handwriting legibility; the greater the legibility, the fewer spelling errors that were made.

Part 2.


The results of Part 2 are not so clear-cut as those that were found in Part 1. The group as a whole had slightly more written errors per student than oral errors, but there was a gender effect. Girls made more oral errors than written, while boys showed the opposite tendency, although more strongly.

Considering the number of errors per word does little to clarify the relationship between handwriting and spelling (Table 3). Although there were more written errors per word than oral errors, the difference is neither statistically nor practically significant.

Computing correlation coefficients between word length and the number of oral or written errors per word resulted in an interesting finding. Word length and the number of oral errors were more highly correlated (r = .71) than were word length and the number of written errors (r = .56). The number of oral and written errors were highly correlated (r = .96).

Discussion.

The results of Part 1 of this study suggest that there is a strong relationship between spelling ability, and handwriting: students with very good handwriting can spell better than their classmates with poor handwriting. [1] As this was not an experimental study, however, it is inappropriate to suggest that good handwriting causes good spelling; or that improving handwriting legibility will bring about a corresponding change in spelling ability. The strength of the association between handwriting and spelling is such that more than mere coincidence appears to link them but until further research of an experimental nature is conducted, there is no justification in declaring 15 that a causal relationship exists between handwriting and spelling.

The outcome of Part 2 of this study contradicts Strickling's (1974) finding that handwriting illegibilities contribute to spelling errors. The difference between the oral and written spelling scores of the participants is the present study was of neither statistical nor practical significance.

One possible reason for the discrepancy between Strickling's research and the present study is that the words on the Morrison-McCall Spelling Scale were less familiar to the participants than were the words used with the South Carolina sample. : It is possible that students have an easier time recognizing and correcting errors in familiar words, even when written poorly, than they do with unfamiliar words.

Word length and oral spelling errors were correlated more highly than were word length and written spelling errors. More than half the variance in oral errors was accounted for by word length, whereas less than one-third the variance in written errors could be attributed to word length. This finding suggests that oral and written spelling involve somewhat different skills, with short-term memory being a more important component of oral spelling, and there is less opportunity, for self-correction.

It is still too soon to assert, as has Petty (1964), that if "handwriting improves, all written work is facilitated with the result of increased benefits to spelling" (p. 6). Handwriting and spelling appear to be related in an educationally relevant way, but until further research is conducted, the strength of the relationship will remain unclear.

References.

Askov, E., Otto, W., and Askov, W. A Decade of Research in Handwriting: Progress and Prospect. Journal of Educational Research, 64 (Nov. 1970), p. 100-111.

Dale, E., and O'Rourke, J. The Living Word. Field Enterprises Education Corp. 1976.

Newland, T. E., An Analytical Study of the Development of Illegibilities in Handwriting from the Lower Grades to Adulthood. Journal of Educational Research, 26, (Dec. 1932) : p. 249-258.

Petty, W. Handwriting and Spelling: Their Current Status in Language Arts Curriculum. Elementary English, 41, (Dec. 1964): p. 839-845, 959.

Strickling, C. The Effect of Handwriting and Related Skills upon the Spelling Score of Above Average and Below Average Readers in the Fifth Grade. Dissertation Abstracts, 1974, 34 (7), p. 3717-A.


Table 1.

Cumulative percentages of students classified according to handwriting legibility and spelling errors

# of spelling Least legibleMost legible
errors123 45

0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
1.4
2.9
4.3
5.8
7.2
12.9
14.3
20.0
30.0
41.5
48.6
55.8
61.5
67.2
68.6
71.5
77.2
85.8
97.2
100.0
4.0
11.0
19.0
24.0
35.5
46.5
59.0
69.5
75.5
80.0
85.5
91.0
94.0
96.0
98.5
99.5
99.5
99.5
100.0
13.6
32.2
46.6
59.9
70.3
78.2
83.1
87.9
93.3
95.2
96.1
96.9
99.2
99.2
99.5
99.8
99.8
100.0
29.9
57.9
80.4
88.8
94.4
97.2
98.1
98.1
99.0
99.0
100.0
52.6
94.7
94.7
94.7
94.7
100.0

Total number
of students.
9020035410719

Table 2.

Number of errors per student

 MeanS.D.
Male (64)oral errors
written errors
3.69
4.11

3.01
3.29
Female (65)oral errors
written errors
3.91
3.85

4.20
3.85
Total (129) oral errors
written errors
3.80
3.98

3.64
3.57

Table 3.

Number of errors, per word

WordOral errors Written errorsTotal

asked
battery
beautiful
bottom
cabinet
ceiling
circle
course
cousin
delivery
friend
hear
money
piece
soap
suit
their
there
voice
wear
20
24
48
13
53
59
26
46
29
35
18
4
6
46
1
5
24
9
10
14
27
18
52
11
56
66
17
53
32
28
11
10
2
48
4
11
25
11
6
25
47
42
100
24
109
125
43
99
61
63
29
14
8
94
5
16
49
20
16
39

Totals4905131003

[1]

Ed. comment:

 It is suggested that poor handwriting is often used as a cover up for inadequate spelling knowledge.

Back to the top.