Comparing Touch and Tilt Interaction using an iPhone Game for

Children

David Furió

1†

, M.-Carmen Juan

1∗

, Ignacio Seguí

, M. José Vicent

and Francisco Abad

Instituto Universitario de Automática e Informática Industrial, Universitat Politècnica de València,

Camino de Vera, s/n, 46022 Valencia, Spain

AIJU, Avda. de la Industria, 23, 03440 Ibi, Alicante, Spain

Keywords:

Mobile Devices, Interaction Preference, Touch Interaction, Tilt Interaction, iPhone.

Abstract:

In this paper, we present a study that compares tilt and touch-screen interaction in a 2D iPhone game for

children. The trials involved 58 children from 8 to 10 years old. The results showed that the children found

touch-screen interaction easier. With regard to engagement and fun, no signiﬁcant differences were found

between the two interaction methods. However, differences between boys and girls were found with the score

given to the game. Another interesting result is that the children had no preferred interaction method.

1 INTRODUCTION

Mobile devices incorporate touch displays and other

hardware features that let the consumers use bodily

and gestural interaction. Touch sensitive screens let

the users have precise control in their actions like

other pointer interaction methods such as a mouse or

a stylus (Schwarten et al., 2008). Touch-based in-

teraction has received positive feedback from users,

which has made possible to improve the user experi-

ence when using mobile devices (Dittenberger et al.,

2009). Bodily and gestural interaction has become

very widespread in the last years, possibly due to pro-

gresses that have made this technology cheaper and

more easily integrated in mobile devices (Kratz et al.,

2011). Bodily interaction that involves device move-

ment, like tilting, has been successfully applied in

gaming scenarios (Kratz et al., 2011). Tilt interaction

method uses the inclination angle of a device. This

kind of interaction let the users have intuitive manip-

ulation since the device is at their hands (Schwarten

et al., 2008).

1.1 Motivation and Objective

Studies about interaction methods that use mobile de-

vices are centred primarily on scrolling web pages,

panning maps or navigating in menus (Schwarten

et al., 2008). These kind of studies are mainly fo-

cused on adults and not on children (Inkpen, 2001).

Furthermore, not many studies examine gender dif-

ferences. For these reasons, we decided to conduct

a study with the aim to investigate children’s interac-

tion experience, ease of use and their preference for

the interaction method in an iPhone 2D game.

In order to obtain a better understanding of chil-

dren’s interaction experience, we tried to answer the

following questions in our study: Will touch interac-

tion be faster and easier to use for children as it is for

adults? Will tilt interaction be funnier for children as

it is for adults? Will gender have a signiﬁcant effect

on the interaction method used?

The paper is organised as follows. Section 2 dis-

cusses related work. Section 3 describes the game.

Section 4 presents the study. Section 5 displays the

results. Finally, in Section 6, a number of conclusions

and areas for future research are identiﬁed.

2 RELATED WORK

In the lasts few years, the advances in mobile tech-

nology has made possible the popularisation of in-

teraction forms that use the accelerometer and the

touch-screen display. Since then, several interaction

studies related to mobile devices comparing tilt or

touch-screen interaction to other interaction methods

have been presented (van Tonder and Wesson, 2010;

Büring et al., 2008; Schwarten et al., 2008; Cho et al.,

2007), but to our knowledge no work that compares

tilt and touch-screen interaction for children has been

submitted.

389

Furió D., Juan M., Seguí I., Vicent M. and Abad F..

Comparing Touch and Tilt Interaction using an iPhone Game for Children.

DOI: 10.5220/0004218403890392

In Proceedings of the International Conference on Computer Graphics Theory and Applications and International Conference on Information

Visualization Theory and Applications (GRAPP-2013), pages 389-392

ISBN: 978-989-8565-46-4

 2013 SCITEPRESS (Science and Technology Publications, Lda.)

In van Tonder and Wesson and Büring, Gerken,

and Reiterer studies map navigation applications were

developed (van Tonder and Wesson, 2010; Büring

et al., 2008). Schwarten et al., and Cho et al., de-

veloped image editing or manipulation applications

(Schwarten et al., 2008; Cho et al., 2007). The re-

sults of these studies showed that tilt interaction pre-

sented some problems such as reﬂections or strained

wrists. Despite this, tilting was considered funnier

and more interesting than other interaction methods.

Touch-screen was considered more precise and easier.

Furthermore, all of these studies were centred on tasks

(e.g., map navigation or image manipulation) that re-

quire to check which interaction method is the most

efﬁcient and usable, leaving the engagement and the

preferences in a second place. But in our case, since

we have developed a game, the engagement and fun

and the preferred interaction methods are also impor-

tant factors to take into consideration.

3 THE GAME

The game was aimed to test which interaction style

(touch-screen or tilt) was preferred and felt more com-

fortable for children. During the game, the children

had to clean a dirty sea (turning the sea’s brown colour

to a blue colour) using a soap that appeared in the cen-

tre of the screen (Figure 1). How they could use the

soap depended on the interaction mode. If the game

was in touch-screen mode, they had to touch-and-drag

the soap around the screen to clean the sea. If the

game was in tilt mode, they had to tilt the device to

move the soap around the screen. There was no time

limit for completing the task and tilt mode had a set

calibration for all the children.

Figure 1: Almost cleaned blue sea.

4 STUDY

Fifty-eight children from 8to 10 years old –with a

mean age of 9.14±0.66– took part in the study: 27

boys (47%) and 31 girls (53%). The children attended

the Summer School of the UPV.

The children who participated in the study were

randomly assigned and counterbalanced to one of two

groups:

• A: Play the game with touch-screen interaction

ﬁrst and then with tilt interaction.

• B: Play the game with tilt interaction ﬁrst and then

with touch-screen interaction.

First, some instructions were given to the children

about how to play the game. Then, the group A played

the game with touch-screen interaction. After com-

pleting the game, they answered the post-game ques-

tionnaire Q1. Next, these children played the game

with tilt interaction and ﬁlled in the ending question-

naire (Q2). The group B did the same but using ﬁrst

tilt interaction and then touch-screen interaction. The

questionnaires were ﬁlled in in the same room where

the activity took place. One person was in the ac-

tivity room in order to clarify doubts to the children.

During the trials, the children used an iPhone 3GS

equipped with an external case to protect the device

against shocks and falls (see Figure 1).

Two questionnaires were used for the validation.

Table 1 shows the relation of questions for each ques-

tionnaire. Answers of questions 1, 3, 4 and 5 followed

a Likert scale that ranged from 1 (strongly disagree)

to 5 (strongly agree). Questions 2, 6 and 7 had cate-

gorical answers (answers in brackets were the possi-

ble choices). And questions 8 and 9 had hand-written

answers. Questionnaires were kept short since they

would be used by children.

The ﬁrst questionnaire (Q1) was composed of

questions 1 to 5 (see Table 1). The questionnaire was

made to collect data about the ease of use and en-

gagement the children had after using an interaction

method.

The second questionnaire (Q2) was composed of

questions 1, 3 and 5 to 9 (see Table 1). Some of the

questions were the same as in Q1, which allowed us to

compare if the order of playing affected the children’s

experience.

5 RESULTS

To measure how much fun the children had play-

ing the game (question 1 in Table 1), we performed

a mixed design ANOVA test to take into consid-

eration several factors simultaneously. The signiﬁ-

cance level was set to 0.05 in all tests. No signiﬁ-

cant differences in any of the factors studied (gender:

GRAPP2013-InternationalConferenceonComputerGraphicsTheoryandApplications

390

Table 1: Numbered questions and their appearance in each

questionnaire. The last two questions are hand written. An-

swers in brackets are a summary of the possible choices.

The rest of answers follow a Likert scale.

# Q1 Q2 Question

1 x x Did you have fun playing the game?

2 x

Would you like to play again? [Yes / No /

Maybe]

3 x x Did you ﬁnd it easy to play?

4 x

Would you like to use this system in other

games?

5 x x Please, score the game from 1 to 10.

6 x

Which interaction method did you ﬁnd it

easier? [Touch-Screen / Tilt]

7 x

Which interaction method do you prefer?

[Touch-Screen / Tilt]

8 x Why?

9 x What did you like the most?

F[1,54]=0.22, p=0.64; grade: F[1,54]=1.61, p=0.21;

method: F[1,54]=3.05, p=0.09; order: F[1,54]=0.52,

p=0.48). The global rating was high (4.59±0.56),

which means that the children had fun playing the

game with both interaction methods.

We performed another mixed design ANOVA test

to measure how easy it was to play the game for the

children (question 3 in Table 1). While the global

rating was high (4.52±0.73), meaning that the chil-

dren found the game easy to play with both inter-

action methods, the score given to touch-screen in-

teraction was higher (4.71±0.53) than tilt interaction

(4.33±0.84). We can see the results of the mixed de-

sign analysis in Table 2, which showed signiﬁcant dif-

ferences in the interaction type.

Table 2: Mixed design ANOVA for the ease of use. N = 58.

The symbol * indicates signiﬁcant differences.

Factor d.f. F p η

Gender 1 0.54 0.46 0.01

Grade 1 0.32 0.57 0.01

Interaction method 1 7.79 0.01 * 0.04

Order 1 0.02 0.88 <0.01

Factor Interactions 1 <1.53 >0.22 <0.01

Furthermore, we asked the children if they would

like to use these interaction systems in other games

(question 4 in Table 1). The score given was high, and

both interaction methods obtained almost the same re-

sult (4.59±0.78 for tilt and 4.59±0.63 for touch). A

multifactorial ANOVA was performed in order to ob-

serve if there were signiﬁcant differences in question

4 responses. No signiﬁcant differences were found

in any of the factors (gender:F[1,50]=1.41, p=0.24;

grade: F[1,50]=0.10, p=0.76; method: F[1,50]<0.01,

p=0.96).

We performed a mixed design ANOVA to measure

the differences in the score given to the game by the

children when playing with touch-screen or tilt inter-

action (question 5 in Table 1). We can see the results

of the mixed design analysis in Table 3, which show

signiﬁcant differences in the gender and interaction

type factors. The girls (9.40±0.88) rated the game

higher than the boys (8.52±1.37) and tilt interaction

obtained higher results (9.12±1.24) than touch-screen

interaction (8.86±1.18).

Table 3: Mixed design ANOVA for the game scores. N =

58. The symbol * indicates signiﬁcant differences.

Factor d.f. F p η

Gender 1 8.23 0.01 * 0.1

Grade 1 0.01 0.91 <0.01

Interaction method 1 4.15 0.05 * 0.02

Order 1 0.25 0.62 <0.01

Factor Interactions 1 <2.20 >0.14 <0.02

A Chi-squared test for question 7 in Table 1, re-

vealed that the preference for the touch-screen or tilt

was not affected by the order of play (χ

[1, N=58]

= 0.07, p = 0.79, Cramer’s V = 0.07). A total of

55.17% of the children in the group A chose touch-

screen interaction and 51.72% of the children in the

group B preferred tilting. We also performed a Chi-

squared test to determine if there were differences in

preference of the interaction types between boys and

girls. The results showed that there were no differ-

ences (χ

[1, N=58] = 1.11, p = 0.29, Cramer’s V =

0.17). Girls preferred slightly more the tilt interaction

(58.06%) while boys favoured marginally the touch-

screen interaction (59.26%).

Regarding the preference for the easiest interac-

tion method (question 6 in Table 1), the children

of both groups favoured touch-screen rather than tilt

(group A: 63.33% and group B: 60.71%). The Chi-

squared test showed no signiﬁcant differences be-

tween the interaction types (χ

[1, N=58] = <0.01, p

= 0.95, Cramer’s V = 0.03). The same test was ap-

plied to determine if there were differences between

boys and girls, but the results revealed no signiﬁcant

differences (χ

[1, N=58] = 0.47, p = 0.49, Cramer’s

V = 0.13). Boys (55.56%) and girls (67.74%) found

easier touch-screen interaction than using tilting.

6 CONCLUSIONS

We have conducted a study with the aim to research

on children’s interaction experience and their pref-

erences about their favourite and easiest interaction

method. Fifty-eight children have participated in the

trials.

ComparingTouchandTiltInteractionusinganiPhoneGameforChildren

391

The results have shown some interesting ﬁndings.

Girls gave a higher score to the game. Girls seem

to prefer to use computers for goal-oriented activities

with meaningful contexts (Kay, 2008). The game de-

veloped provided a goal with a meaningful context

(clean the dirty sea). This might have been the reason

why the girls scored higher the game. With regard to

the results about ease of use, we could see that both

groups considered touch-screen interaction easier and

more precise than tilt interaction. This is encouraging

because it shows that children could adapt to differ-

ent interaction types without major problems. With

regard to engagement and fun, the results revealed

that the children had similar enjoyment when play-

ing with tilt and touch-screen interaction. With regard

to the general preferences about the game, the analy-

ses showed that the children did not favour one inter-

action over the other. Subjects also stated that they

would like to use these interaction systems in other

games.

When comparing our study to other studies with

adult people, we can observe similar trends with re-

gard to the ease of use, where touch-screen interac-

tion was found easer and more precise (Schwarten

et al., 2008; van Tonder and Wesson, 2010). In con-

trast, there are differences with regard to the engage-

ment and fun. While, in our study, children consid-

ered equally fun playing with touch-screen or tilt in-

teraction, adults usually found funnier using tilt inter-

action (Schwarten et al., 2008; Cho et al., 2007). It

is important to note that in the studies of (Schwarten

et al., 2008) and (Cho et al., 2007) the tasks performed

were image manipulation and image navigation, re-

spectively. In our case, the task performed was to play

a game. Thus, it is possible that playing an engag-

ing task (like a game) made that the children found

the two interaction methods fun. A future study must

be conducted in order to see how the children would

rate the interaction types in an application that is not

a game.

With regard to future work, allowing them to cal-

ibrate the accelerometer when using tilt interaction

could enhance the experience. The game could also

be improved by adding different levels where obsta-

cles appear in the middle of the screen making the

experience more difﬁcult. We could also add other

interaction methods to compare, like the use of but-

tons. In this study, using the device with or without

an external case was not tested. Thus, it would be in-

teresting to perform a comparison between using the

device with the external case and using it without the

external case, in order to observe whether or not the

external case actually helps the children move the de-

vice.

ACKNOWLEDGEMENTS

This work was funded by the Spanish APRENDRA

project (TIN2009-14319-C02). For their contribu-

tions, we would like to thank the following: the peo-

ple who helped in the development and validation; the

Summer School of the UPV; the children who partic-

ipated in this study; and the ETSInf for letting us use

its facilities during the testing phase.

REFERENCES

Büring, T., Gerken, J., and Reiterer, H. (2008). Zoom

interaction design for pen-operated portable devices.

International Journal of Human-Computer Studies,

66(8):605–627.

Cho, S. J., Choi, C., Sung, Y., Lee, K., Kim, Y. B., and

Murray-Smith, R. (2007). Dynamics of tilt-based

browsing on mobile devices. In CHI 07 extended ab-

stracts on Human factors in computing systems CHI

07, page 1947. ACM Press.

Dittenberger, S., Geven, A., Tscheligi, M., and Mayer, M.

(2009). Touch based interaction using a three dis-

play interface design. Proceedings of the 11th In-

ternational Conference on HumanComputer Interac-

tion with Mobile Devices and Services MobileHCI 09,

page 1.

Inkpen, K. M. (2001). Drag-and-drop versus point-

and-click mouse interaction styles for children.

ACM Transactions on Computer-Human Interaction,

8(1):1–33.

Kay, R. (2008). Exploring Gender Differences in

Computer-Related Behaviour: Past, Present, and Fu-

ture. In Kidd, T. T. and Chen, I., editors, Social Infor-

mation Technology: Connecting Society and Cultural

Issues, pages 12–30. IGI Global.

Kratz, S., Rohs, M., Wolf, K., Müller, J., Wilhelm, M., Jo-

hansson, C., Tholander, J., and Laaksolahti, J. (2011).

Body, movement, gesture & tactility in interaction

with mobile devices. Proceedings of the 13th In-

ternational Conference on Human Computer Interac-

tion with Mobile Devices and Services MobileHCI 11,

page 757.

Schwarten, L., Walther-Franks, B., Grimmer, C., and Feige,

S. (2008). A comparison of motion and keypad in-

teraction for ﬁne manipulation on mobile devices. In

Proceeding BCS-HCI ’08 Proceedings of the 22nd

British HCI Group Annual Conference on People

and Computers: Culture, Creativity, Interaction, vol-

ume 2, pages 93–103.

van Tonder, B. and Wesson, J. (2010). Is tilt interaction

better than keypad interaction for mobile map-based

applications? In Proceedings of the 2010 Annual

Research Conference of the South African Institute of

Computer Scientists and Information Technologists on

- SAICSIT ’10, pages 322–331, New York, New York,

USA. ACM Press.

GRAPP2013-InternationalConferenceonComputerGraphicsTheoryandApplications

392