Typographic Design for Children

The songs the letters sing: typography and children’s reading

Sue Walker

This booklet arises from the Typographic Design for Children project based in the Department of Typography & Graphic Communication at The University of Reading.

The project began in 1999 and is due for completion in 2005. Further information can be found on its website: www.kidstype.org

The project team is Sue Walker (director), Linda Reynolds (co-director), Alison Duncan (research assistant), Nicola Robson (research student) and Nadja Guggi (research assistant). The team also acknowledges the work of Caroline Archer and Angela Redfern, who carried out some of the research in schools and data analysis in the early stages of the project. We would like to thank teachers and children at Redlands Primary School and Aldryngton Primary School in Reading for their help with the testing. Oxford University Press kindly supplied text and illustrations for the test materials, which were printed by the Design and Print Unit in the Department of Typography & Graphic Communication at The University of Reading.

Typographic Design for Children

is funded by

The AHRC funds postgraduate training and research in the arts and humanities, from archaeology and English literature to design and dance. The quality and range of research supported not only provides social and cultural benefits but also contributues to the economic success of the UK. For further information on the AHRC, please see their web site www.ahrc.ac.uk

© Typographic Design for Children 2005

ISBN 07049 98467

National Centre for Language and Literacy The University of Reading Bulmershe Court Reading RG6 1HY

Designed and produced by Text Matters

Typographic Design for Children 3

Opinion and research about typography in children’s reading books 5

Are sans serif or serif types easier to read? 7

Are a and g more difficult than a and g? 9

What do children think about script and informal typefaces? 10

Do children need ‘special’ typefaces? 12

Space between letters and words 14

Space between the lines 17

Ways that typography can help children’s reading 19

References 23

Contents

3

Does typography have a significant effect on children’s reading? Does it affect their motivation to read? Are serif or sans serif types easier or more difficult for children to read? Do children find infant characters (eg variants of ‘a’ and ‘g’) easier or more difficult to read? Does more or less space between letters, words and lines help or hinder children’s reading?

These are the questions we wanted to find answers to when we embarked upon a series of performance and preference tests to consider the extent to which the typographic design of reading books affects reading performance and motivation.

We worked with editors and designers at Oxford University Press (OUP) to identify a suitable base text to use for our testing with beginner readers. A sheepless night (Oxford, 1994), written by Geraldine McCaughrean and illustrated by Mike Spoor, is part of the Oxford Literacy Web designed to fit the National Literacy Strategy requirements in primary schools. A sheepless night is part of the Oxford Literacy Web’s Fiction Strand (‘fun-packed stories every child will love’) and is aimed at children about six years old, likely to be in years 1 and 2 in primary schools. This relates to Individualised reading stage 7 (Reading: Reading and Language Information Centre, 2001) which is National Curriculum working within level 1, age six to seven. The 24-page book fitted the project requirements well: each page contained an illustration and some text (minimum 2 lines, maximum 7), and as such offered flexibility for changing some of the typographic parameters for our testing. It provided the story, illustrations and format for the material for each of the tests. For each test the pages were printed in full colour to match the original books published by OUP.

Spread from A sheepless night (Oxford, 1994). The original version was set in a serif typeface, OUP Baskerville.

Typographic Design for Children

4

We listened to children reading in a classroom using high quality test materials that were realistic in terms of children’s expectations in this context. The children, from primary schools in Reading, Berkshire, were used to reading to adults who were not their class teacher, so the situation was not unusual or intimidating.

For each of the tests, we recorded the children’s reading and then undertook miscue analysis. Miscue analysis is one of the ways that teachers build up information about a child as reader, though it has recently fallen out of favour with the re-introduction of phonic approaches to teaching reading. Some researchers (eg Moon, 1979) have used miscue analysis as a means of finding out about the text as opposed to the reader.

The children’s errors were categorised, counted and statistically analysed. Full details of this testing can be seen in Walker and Reynolds (2002/03) and Reynolds and Walker (2004). We also asked children what they thought about the different typefaces and spacing variants that they saw. Could they ‘spot the difference in the way the writing looks’ when comparing typefaces? Which of the typefaces did they like best? Which did they find hard to read?

Our results from the miscue analysis were not statistically significant, but the comments the children made about the typographic variations they saw suggested that typography may have considerable effect on their motivation to choose and engage with a book. It is these comments through which the children discussed their understanding of letterforms and space that provide the main focus of this booklet.

5

Teachers tend to have strong views about typefaces in children’s books, in particular about the use of serif or sans serif type, and the use of infant characters. One such view is that sans serif types (with infant characters) are most suitable for beginner readers, and publishers responding to this typeset books for beginner readers in large sans serif types, such as Frutiger and Helvetica, often with specially-designed infant characters.

A survey of current (2004) reading schemes on display in the National Centre for Language and Literacy reveals that this is indeed the case. Versions of Helvetica, Frutiger and to a lesser extent Gill Sans, continue to be widely used in the early stages of reading schemes. The survey included the following popular reading schemes: Longman’s Story Street books, OUP’s Oxford Reading Tree, Heinemann’s Storyworlds, Ginn’s All Aboard books, Nelson Thornes’ New PM Story Books, and Collins’ Pathways. All of these schemes use infant a’s and g’s and alternative numbers in the early stages of the scheme.

In practice in schools children often read a wide range of books, not only those from a particular reading scheme. Many primary schools offer books produced by ‘trade’ publishers as well as ‘educational’ ones. When this is the case children are presented with many different typefaces, both serif and sans serif with and without infant characters, because trade publishers are not restricted in their choices to typefaces that are thought to be more likely to sell books in schools.

In 1984 Raban conducted a survey amongst teachers to find out what features (including typographic ones) they thought were important when choosing books for young children. Two-thirds of the 271 teachers in her sample favoured the use of sans serif type throughout the infant school because they thought that clean, clear shapes corresponded closely with children’s handwriting. These teachers also preferred infant a’s and g’s in books for five- and six-year old children, but thought this was less important for seven-year olds and above. Teachers in a study by Coghill (1980) also thought that non-infant a’s and g’s would ‘cause difficulty’ for children. But Coghill also found that children were not confused by alternative letter shapes:

Children are less sensitive to changes in the ‘look’ of letters, words and sentences than one might imagine them to be. The fact that children are able to use their newly acquired reading knowledge to decipher a variety of letter forms indicates that children are more adaptable than many teachers believe. (p.260)

Sassoon (1993, p.161) also supports this view:

… there is no proof at all to support the proposition that most children find it difficult to discriminate between simple written and printed forms, even forms of the ‘g’ and ‘a’. On the contrary, in real life situations, our children are, from an early age, bombarded by television graphics and other advertising. Apart from those with severe perceptual problems, children appear to assimilate the different forms quite happily.

Opinion and research about typography in children’s reading books

‘Infant character’ is used to describe letters designed with the perceived needs of children in mind. Sometimes letters are redrawn to look like handwritten forms and sometimes they are specially drawn to be clearly distinguished from similar-looking letters.

6

The teachers in Raban’s survey ranked spacing – of words, letters and lines – lower than size of print or style of typeface when choosing books for children. However, these teachers thought that for five- and six-year olds word spacing was more important than line and letter spacing, but for seven-year olds that space between lines was more important than word and letter spacing. These teachers appeared to associate generous word spacing with what was appropriate for young children.

There has been very little formal research about typography in children’s books. Much of what there is has been undertaken by psychologists (see Watts and Nisbet, 1974 and Venezky, 1984), often with very little awareness both of design issues and of classroom practices.

More recently, and specifically on word spacing, Cooper-Tomkins (1994) surveyed word spacing in contemporary reading schemes. This small study of mainly sans serif typefaces showed a wide variety of word and letter spacing practices. It indicated that many children’s books are very tightly letter and word spaced – a practice criticised by Yule (1988) and Sassoon (1993) – and that some are so widely spaced that it may make it difficult for children to perceive lines of text at all (Walker, 1992). Sassoon’s work (1993) explored children’s preferences for typeface and word spacing. She concluded that where spacing is concerned, presenting findings as an average or norm is not always useful as children at different levels of reading have different requirements for spacing.

Yule (1988) noted that the line spacing in children’s books is usually wide enough to prevent most children’s eyes slipping from line to line in reading. Hughes and Wilkins (2002) have found generous line spacing could benefit children’s reading. They investigated the effect of type size and spacing on the ability to read at a distance, in order to help with the design of text in children’s ‘big books’. They varied both letter spacing and line spacing in relation to type size, testing the visual acuity and reading speed of 200 children of six to twelve years of age. They concluded that ‘children’s reading would benefit by increasing the font size of the text and by expanding the spacing horizontally and vertically’ (p.225).

7

One of our tests compared Century (a serif type) and Gill Sans (a sans serif type) with and without infant characters.

Century and Gill Sans were chosen as typefaces for this because both typefaces have reasonably generous ascenders, and the length of the descenders is similar in relation to their x-heights; both have reasonably well-designed infant characters, and both are tried and tested reading scheme typefaces. Care was taken so that the type was the same appearing size on all material; that the line breaks were the same on all the specimen pages, and were broken in accordance with the principles outlined in the research carried out by Raban (1982); and that the line feed appeared the same on each of the specimen pages. Four specimen books were set in Century and Gill Sans, each with infant characters and without. An example of the type used in each of the books is shown below.

Are sans serif or serif types easier to read?

Examples of type used in Test 1: •Century•CenturyEducational•GillSans•GillSchoolbook

8

The children were asked if they could ‘spot the difference in the way the writing looks’ when comparing typefaces, and then which of the writing styles they liked best, and which they found harder to read.

Fifteen of the 24 children in our sample noticed or thought they noticed differences of some kind, including size, weight, addition of serifs, and between particular characters. Those who noticed differences in size thought that Gill and Gill Schoolbook looked larger than Century and Century Educational. The versions of Gill were thought to be heavier than the versions of Century.

Some children noticed that Century had serifs and that Gill did not. One child commented:

The [Century] capital ‘W’ has little lines on top. The [Gill] capital ‘W’ doesn’t …’, and another: ‘The capital ‘T’s are different. [Century Educational] has two little points on, [Gill Schoolbook] doesn’t.

Of the 24 children, 8 prefered Gill Sans, and 3 Gill Schoolbook. Only 2 liked Century, and 3 Century Educational. However 8 children expressed no preference.

The children said the following about the Gill typefaces:

• thelettersstandoutandaremuchbiggertoread

• ithasn’tgotanylittlelines.Idon’tlikethelittlelines–it spoils it

• nicethickletters

• biggerletters

• lookseasytometoread

And the following about Century:

• youcanreaditproperly

• itistheeasiesttoread

• itisquiteeasytoread,anditlooksdifferent

Differences noticed by children: in each illustration Century is followed by Gill.

Century’s capital W has ‘little lines on top’

Century’s capital T has ‘two little points on’

9

As well as seeing whether children noticed differences in and had preferences for serif and sans serif type, our first test also considered children’s views about infant characters. Century Educational and Gill Schoolbook have both been designed with a number of specially modified characters that take account of teachers’ preferences. Most of the shape changes are a result of making the characters more like handwritten letterforms. Some changes are to distinguish between letters that may be confused as, for example, capital ‘i’ and figure one.

Most children in our study were well aware that there were different forms of a and g, and some even made the point that a is what we write and a is what we read. Some commented favourably about what they perceived as the ‘normal’ ‘g’ used in Gill Schoolbook. One child made particularly detailed comments:

Century and Gill are both kind of the same. No, they’re not. You see ‘reading’? Look at that ‘a’ then look at that ‘a’ – they are not the same looking. Look at that ‘g’ and then look at that ‘g’. I like ‘g’ best. If you look at the ‘y’ and you look at that ‘y’ they’re not the same. The ‘y’ is like a ‘u’ but it goes down and is a bit curvy. But if you look at the [Century] ‘y’ it goes straight down.

Others thought that a and g were ‘harder’ than a and g, but this did not affect their reading performance in our tests.

Not all typefaces with infant characters are helpful for children’s reading. If infant characters are used it is important that there is clear differentiation between characters such as o, a and g. This is the case in Gill Schoolbook, for example, but not in Avant Garde Gothic or Helvetica.

Are a and g more difficult than a and g?

Century a g y l I 4 9 1Century Educational a g y l I 4 9 1Gill Sans a g y l I 4 9 1Gill Schoolbook a g y l I 4 9 1

‘this is what we write’

‘this is what we read’

Infant modifications in Century Educational and Gill Schoolbook.

In some typefaces, such as Avant Garde Gothic, there is very little differentiation between some letter shapes

10

A second set of tests compared Century and Gill with two informal-looking typefaces, Sassoon Primary Infant and Flora. The specimen books were set in Century Educational, Gill Schoolbook, Flora and Sassoon Primary Infant. Supplementary test material was also produced as double-page spreads set in Comic Sans, French Script, Fabula, Lucida Handwriting and Sand.

Again, we asked the children whether they could ‘spot the difference’, and about their preferences. The children noticed differences in features of individual letters, and in the qualities of the setting as a whole, such as size and texture.

• In comparing Flora with Sassoon, for example, one child observed that in Sassoon the f’s have a ‘curl’ and in Flora they are ‘just lines’. Another difference noticed between these typefaces was that Sassoon has lines on the capital I and Flora does not.

• A particularly distinctive feature noticed by a quarter of the children in the sample was that Flora’s quotation marks had no ‘circle’ and were just lines. The Century quotation marks were described as ‘sixes’.

• Half the children in the sample commented on the qualities of the typefaces more generally. Flora, for example, was thought to be ‘curvy’, ‘fancy’ and ‘grown up’. The comments made by the children suggested that Flora has distinctive or unfamiliar qualities that make it stand out from others. Several children associated particular circumstances of use with certain typefaces. For example, Century was described as ‘normal, and like a computer’ The same child thought that Flora was also ‘like something on a computer’ and that Sassoon was ‘like someone’s written it – flowing’.

What do children think about script and informal typefaces?

Example of type used in Test 2:• Flora•SassoonPrimaryInfant

In this test children compared these typefaces with Century Educational and Gill Schoolbook, which are illustrated on page 7.

Flora’squotationmarkshave no ‘circle’, Century’s look like sixes

In Sassoon Primary Infant thef ’shavea‘curl’;inFlorathey are ‘just lines’

The Sassoon Primary Infant ‘I’ has ‘lines on it’

11

Some of the phrases children used to describe the typefaces are shown above.

Some children used words that suggested certain typefaces may affect motivation: ‘brighter’ and ‘faster’, for example, are more positive than ‘babyish’, ‘funny’ and ‘silly’.

The children also had views on the appropriateness of particular typefaces: some found that even when they thought a typeface was attractive, they thought it was not suitable for use in a reading book. For example, one girl said of French Script: ‘I can’t read it very well, but I like the look of it’. Some features of letters were thought to be distracting if they were too noticeable, as suggested by the words ‘curly’, ‘loopy’, ‘wiggly writing’.

Associations can be strong, and there seems to be some comfort for young children in reading things that look familiar or ‘normal’.

Comic Sans• ‘babywriting’• ‘likewritinginacomic’

Shep was the sheepdog, but he did not like sheep.He said they were silly and boring.He said they did nothing all day but eat grass and say baa baa.

2 3fabula fabula

He said, “I wish I could look after lions, or tigers, or monkeys – not silly sheep.”

Shep was the sheepdog, but he did not like sheep.

He said they were silly and boring.

He said they did nothing all day but

eat grass and say baa baa.

2 3Lucida handwriting Lucida handwrtiting

He said, “I wish I could look after lions,

or tigers, or monkeys – not silly sheep.”

Shep was the sheepdog, but he did not like sheep.

He said they were silly and boring.

He said they did nothing all day but

eat grass and say baa baa.

2 3Sand Sand

He said, “I wish I could look after lions,

or tigers, or monkeys – not silly sheep.”

FrenchScript• ‘foreign’• ‘girly’• ‘liketheoldendays’• ‘birthdaycardwriting’

Fabula• ‘realwriting’• ‘likeanordinarybook’

Lucida Handwriting• ‘oldfashioned’• ‘special’

Sand• ‘easybecausenot

joined up’

12

Our work to date has shown that children read serif and sans serif types equally well in a situation where typographic variables are carefully introduced, and that they are sensitive to stylistic qualities that typefaces have. In our tests we included two typefaces designed with the needs of children in mind: Sassoon Primary Infant and Fabula. In the performance tests Sassoon performed no better or worse than Gill, Century or Flora, and in the preference tests neither Sassoon nor Fabula were thought by the children to be particularly helpful or notable.

Nevertheless, both these typefaces were designed specially for young children, and some characteristics, such as long ascenders and a friendly ‘feel’ may be important qualities to look for in other typefaces. Type designers and typographers are also likely to know which typeface characteristics may be unhelpful for children’s reading. Typefaces with very short ascenders and descenders is one example, and another is those with very round or geometric letterforms which are likely to reinforce similarities between confusible characters such as a, g and o.

Sassoon Primary

Shep was the sheepdog, but he did not like sheep. He said they were silly and boring.Sassoon Primary, designed by Rosemary Sassoon in the 1980s, was a conscious attempt to design a typeface with characteristics that would be helpful to children’s reading. Sassoon elicited the views of around 100 children in her work, half of whom were eight- to thirteen-year olds with special needs, and half were eight-year olds in mainstream education. She found that children could be very ‘firm and verbose about their opinions’ on typefaces, and that children at different levels of attainment had different requirements for typography. Many children favoured typefaces that included characteristics of handwriting such as slant and other cursive qualities. Sassoon Primary Infant has the following characteristics:

• sans serif terminals for ascending strokes, and flowing ‘flick up’ exit strokes on the baseline.

• alternative letterforms for G, I, J, f, k, q, &, 4, and 9 (as well as Welsh accented letters)

• long ascenders and descenders

In our tests children described Sassoon Primary Infant as: ‘flowing’, ‘small’, ‘light’ and ‘like someone’s written it’.

Do children need ‘special’ typefaces?

Sassoon Primary Infant is one of the Sassoon ‘family’ of typefaces. It is an upright version of the original Sassoon Primary. You can find out more at www.clubtype.co.uk

long ascenders and descenders

alternative character shapes for some letters

flick up exit strokes on the baseline

dog

13

Fabula

Shep was the sheepdog, but he did not like sheep. He said they were silly and boring.Fabula was designed in 1999 as a screen font for children as part of the Fabula Project, which produced software to enable children and teachers to produce bilingual story books.

This typeface has a number of characteristics that the project team held to be important for children’s reading:

• long ascenders and descenders to help identify the word shape

• an informal ‘feel’

− rounded stroke ends

− a rounded ‘e’

• a clear distinction between characters that might be easily confused

• the font uses an alternative infant ‘g’, but not a corresponding infant ‘a’ because we wanted to make as much differentiation between the sometimes confused ‘a’ and ‘o’.

• clear distinction is made between small letter ‘l’, capital ‘i’ and figure ‘1’

Fabula also contains all the accented characters necessary for a wide range of languages including Welsh, Frisian and Catalan.

Even though in its current form Fabula is designed for screen use it was one of the typefaces we included in our tests. Words children used to describe it included: ‘clear, so you can see it properly’, ‘thick, fat’, ‘normal’, ‘like an ordinary book’.

One of the dangers with types designed especially for children is that taking into account qualities that might make them ‘friendly’, such as rounded terminal strokes, or the addition of ‘flicks’, can mean that they do not have the overall appearance of a conventional typeface. If one of the purposes of teaching children to read is to familiarise them with the typographic conventions of reading, then specially-designed typefaces may not be the answer. Selecting established typefaces that have generous ascenders and descenders, clear distinction between characters that are sometimes confused, and no quirky or unusual characters might be an equally good approach to typeface selection.

clear distinction between l, I and 1

infant ‘g’

rounded stroke end

long ascenders and descenders dog

14

Typography is about much more than typefaces. The way in which typefaces are used – the spaces between letters, words and lines – is equally important. We showed the children four letter spacing and four word spacing variants.

More children noticed differences in letter spacing than differences in word spacing. Of the 24 children in the letter-spacing sample, 22 could see differences. 17 children thought that the very tight setting looked the most difficult to read. They thought that the tight spacing made the type look darker, thicker or smaller, and described it:

• ‘mostdifficult,squashedtogether,youcan’tseewhatthewords say’

• ‘youcan’tseethewritingverywell’

• ‘mostdifficult...stucktogether’

Space between letters and words

Letter spacing variants:• tight•normal•wide• verywide

15

Perceptions about word spacing were less conclusive. Of the 24 children who talked about their preferences only 13 could see any differences. Some found it difficult to describe their preference in terms of the space between the words, using terms such as ‘bigger’, ‘darker’, ‘thicker’ and ‘blacker’ to describe the appearance of the type on the pages with narrow word spacing, and ‘smaller’ or ‘thinner’ to describe type on wide word spacing pages.

Three children (of the 13 who could see any differences) thought the narrow spacing was the easiest to read describing it as:

•‘smallspace,bigwriting...easiertoread’

Five thought that the very wide setting was the easiest and appeared to appreciate the ‘wide apart’ appearance of the text.

Word spacing variants:•narrow•normal•wide• verywide

16

Children’s comments about letter and word spacing suggest that smaller or greater amounts affect their views about size and texture, and level of difficulty.

The normal and wide settings used in our test material for word spacing were thought to be ‘normal’, whereas the narrow spacing was described as ‘funny’ and ‘not how words should look’.

Close letter and word spacing may make the text appear more difficult and more confusing. With narrow word spacing some children commented that ‘it just looks like a massive word’ or ‘you have to read a lot of words together’.

Some children thought that wider spacing was helpful; the type was ‘not scrambled up’ and ‘leaves spaces so you can spell the word’.

We recorded the words that children used to describe the typographic variants they saw. It confirmed that they were aware of differences and were articulate in their views. Such awareness may influence their motivation to read.

Narrow Very wide

Word spacing squashed up

tight

small/short spaces

not as many spaces

big writing

black letters

close together

words are all near

close together lines

all one word

all the words are running together

wide

words are far apart

long lines

has far apart lines

small writing

big/long spaces

lots of spaces

all spaced out

spaces between the words

Tight Normal Wide Very wide

Letter spacing squashed

close together

all stuck together

letters get confused

dark

thick

straight

small

black

tiny spaces

doesn’t have a gap between each letter

looks funny, not how words should look

easy to see

not squashed

not joined up

not close together

bit thick

looks normal

sort of spread out but not too spread out

has little bit of space

the way words should be

big

easy to see

not squashed

not scrambled up

thin writing

normal

nearly normal

far apart and separated

has a bit of a gap, the gaps are wide enough, I am used to reading things like that

you can see the letters easily

far apart

big writing

thin writing

spread/spaced out

stretched out

looks wide

big spaces, gaps

not together

separated

you can see it properly

leaves spaces so you can spell the word

looks like there is more writing on the page

Words children used to describe variations in letter spacing and word spacing

17

Our final tests asked children what they thought about variations in space between the lines. Once again A sheepless night was modified to produce four variants set in 19 point Century Educational with baseline to baseline increments of 17, 21, 30 and 39 points.

Space between the lines

4 5Close lf Close lf

But Shep was a good dog. So every day he sat with his nose on hispaws and he counted sheep. Then one day he fell fast asleep.

When he woke up, it was dark. The moon was high and white in the sky .He could see the whole field in the moonlight.But he could not see any sheep – not even one!

4 5V close lf V close lf

But Shep was a good dog. So every day he sat with his nose on hispaws and he counted sheep. Then one day he fell fast asleep.

When he woke up, it was dark. The moon was high and white in the sky .He could see the whole field in the moonlight.But he could not see any sheep – not even one!

4 5Widest lf Widest lf

But Shep was a good dog.

So every day he sat with his nose on his

paws and he counted sheep.

Then one day he fell fast asleep.

When he woke up, it was dark.

The moon was high and white in the sky .

He could see the whole field in the moonlight.

But he could not see any sheep – not even one!

4 5CentInf–T1 CentInf–T1

But Shep was a good dog.

So every day he sat with his nose on his

paws and he counted sheep.

Then one day he fell fast asleep.

When he woke up, it was dark.

The moon was high and white in the sky .

He could see the whole field in the moonlight.

But he could not see any sheep – not even one!

18

When the children were asked which version they found easiest to read, there was a tendency for those who answered to prefer the wide and normal spacings, usually for reasons relating to the spacing or the apparent size of the type.

Opinions on the hardest version were more evenly divided, but when the children were asked which version they would like to take home, 16 of the 24 chose either the wide or normal spacing. They gave a variety of reasons, six basing their choice explicitly on aspects of spacing and three on the perceived size of the type.

One child could distinguish between all four versions, describing the wide spacing as having ‘big spaces’, the normal as having ‘middle sized spaces’, the close as having ‘little spaces’ and the very close as having ‘line after line’ of text, referring to the lack of space.

Very close Close Normal Wide

Line spacing close together

dark

more writing

line after line

words close together

squashed

near to each other

no spaces/gaps

don’t have to search for which line you’re on

smaller writing – can’t see the words

close

more writing

little sized spaces

smaller writing

together

not spreaded

blurry

looks real

spreaded

bigger writing, you can tell which words are there

spaces in between the sentences

light

not close

big spaces

spaced/spread out

bigger writing

more clear

lots of lines muddled up, start reading the same line again

attractive, good to read

has holes to tell you which line you’re on

Words children used to describe variations in line spacing

19

Our discussions with children indicated that they can be very tolerant of a wide range of typographic variants. In practice, some of the extreme versions of our test material, with very close letter, word and line spacing, would never be used. However, children’s ability to cope with such extremes, albeit in a test situation, suggests that focusing on very particular issues, such as whether serif or sans serif type is ‘best’, or whether ‘a’s and ‘g’s should be single or double-storey, may not be as relevant as once thought.

The following points are the most important to emerge from our discussions with children:

• Motivation to read can be affected by typefaces thought to be inappropriate for a reading book, or too distinctive. Children in our sample used the word ‘normal’ to describe Century, Gill and Fabula, suggesting they thought these typefaces were suitable.

• Many children spoke of ‘reading a’s’ and ‘writing a’s’ suggesting they were well aware of the difference in shape between single and double storey forms. Though some children perceived a and g as ‘harder’ than a and g, it did not affect their motivation to read.

• Letter,wordorlinespacingthatwasverynarrowwasthoughtby many children in our sample to make text seem more difficult.

• Our children presented wide-ranging and articulate views which suggest that it is unlikely that one style/kind of typography will suit every child. Discussion of typographic attributes – the typeface and the space between the letters, words and lines – should be a significant element in helping children choose a book.

In our discusssions words such as ‘normal’ and ‘easy’ to describe both typeface and spacing configurations suggested to us that certain forms of typography might be more suitable than others if you are trying to motivate children to read.

The following annotated examples show the specimen pages from our test material that the children in our sample described as ‘normal’ and easy’. They did, however, have much to say about the more extreme variants in the sample and this reinforces our view that encouraging children to read should include discussion about letterforms and their spacing. Children’s views about typography are just as valid as those of teachers and publishers.

The examples on the next pages show the kind of typography that our discussions with children suggest is helpful for beginner readers around the age of six, working within key stage 1.

Ways that typography can help children’s reading

20

Shep was the sheepdog, but he did not like sheep.He said they were silly and boring.He said they did nothing all day but eat grass and say baa baa.

2 3fabula fabula

He said, “I wish I could look after lions, or tigers, or monkeys – not silly sheep.”

Fabula• straight•big,thick,fat,ª curly•normal• clearsoyoucanseeit

properly

Century•neat,plain• straight,normal• likeacomputer• looksliketheymadeit

properly

Gill•big,thick,fat•dark,black,plain•normalletters•baby’swriting• thesameasIreadinall

the other books

The children in our sample used the words ‘normal’ and ‘easy’ to describe the typefaces illustrated below.

However, many typefaces are suitable for children’s reading, provided that they are well spaced. They should have:

• relativelylongascendersanddescenders

• cleardistinctionbetweencharactersthatmightbe confused

•noquirkyorunusuallettershapes

Typeface

21

Space between words and letters

‘Normal’ and ‘wide’ word spacing were what our children expected to see. Given that closely spaced letter or word spacing was perceived by some children as making the text more difficult to read, there may be some advantage in choosing books that have relatively wide word spacing as shown below.

Similarly, with letter spacing: ‘normal’ and ‘wide’ were perceived as meeting their expectations. One problem that can occur in words that are letter-spaced is that they can appear very fragmented, and to compensate for this word spacing and line spacing must be adjusted accordingly. The ‘wide’ letter spacing shown below is the maximum we would recommend.

‘Normal’ word spacing

‘Wide’ word spacing

‘Wide’ letter spacing

22

4 5CentInf–T1 CentInf–T1

But Shep was a good dog.

So every day he sat with his nose on his

paws and he counted sheep.

Then one day he fell fast asleep.

When he woke up, it was dark.

The moon was high and white in the sky .

He could see the whole field in the moonlight.

But he could not see any sheep – not even one!

4 5Widest lf Widest lf

But Shep was a good dog.

So every day he sat with his nose on his

paws and he counted sheep.

Then one day he fell fast asleep.

When he woke up, it was dark.

The moon was high and white in the sky .

He could see the whole field in the moonlight.

But he could not see any sheep – not even one!

The spacing in this example is typical of that seen in contemporary reading schemes for beginning readers; it would appear that current practice is generally appropriate but that no harm would be done by using slightly more generous spacing.

Space between the lines is one of the most important things to think about when you are choosing books for children.

In our discussions with children the examples shown on this page were the ones most favourably received.

Space between lines

23

Coghill, V. (1980). Can children read familiar words set in unfamiliar type? Information Design Journal, 1, 4, pp.254–60.

Cooper-Tomkins, G. (1994) An investigation into the effects of word spacing on the fluency and accuracy of novice readers. Unpublished undergraduate dissertation, Department of Typography & Graphic Communication, The University of Reading.

Hughes, L. and Wilkins, A. (2000). Typography in children’s reading schemes may be suboptimal: evidence from measures of reading rate. Journal of Research in Reading, 23, 3, pp.314–24.

Hughes, L. and Wilkins, A.J. (2002). Reading at a distance: Implications for the design of text in children’s big books. British Journal of Educational Psychology, 72, 2, pp.213–26.

Moon, C. (1979). Categorization of miscues arising from textual weakness. In D. Thackray (ed) Growth in reading: proceedings of the fifteenth annual course and conference of the United Kingdom Reading Association, Nene College, Northampton, 1978. London: Ward Lock Educational, pp.135–46.

Raban, B.(1982). Text display effects on the fluency of young readers. Journal of Reading Research, 5, pp.7–28.

Raban, B. (1984). Survey of teachers’ opinions: children’s books and handwriting styles. In D. Dennis (ed) Reading: meeting children’s special needs. London: Heinemann, pp.123–29.

Reynolds, L. and Walker, S. (2004). ‘You can’t see what the words say’: word spacing and letter spacing in children’s reading books. Journal of Research in Reading, 27, 1, pp.87–98.

Sassoon, R. (1993). Through the eyes of a child: perception and type design. In R. Sassoon (ed) Computers and typography. Oxford: Intellect Books, pp.150–77.

Venezky, R. L. (1984) The history of reading research. In D. P. Pearson, R. Barr, M. L. Kamil and P. Mosenthal Handbook of reading research. London: Longman, vol 1 pp.3–38.

Walker, S. (1992). How it looks: a teacher’s guide to typography in children’s books. Reading: Reading and Language Information Centre.

Walker, S. and Reynolds, L. (2002/03). Serifs, sans serifs and infant characters in children’s reading books. Information Design Journal, 11, 2/3, pp.106–22.

Watts, L. and Nisbet, J. (1974). Legibility in children’s books: a review of research. Slough: NFER Publishing Company Ltd.

Yule, V. (1988). The design of print for children: sales appeal and user appeal. Reading, 22, pp.96–105.

Zachrisson, B. (1965). Studies in the legibility of printed text. Stockholm: Almqvist and Wiksell.

References

The title of this booklet, ‘The songs the letters sing’, is borrowed from a well-known reading scheme first published in the 1920s by the Grant Educational Company. Illustrated by Margaret Tarrant, the series was extremely popular and was used in schools until the 1950s. The early books in the series are set in a distinctive seriffed typeface which, unusually for its time, has infant a’s and g’s.