veronika karnowski a,⇑

8/12/2019 Veronika Karnowski a,

1/10

When lifestyle becomes behavior: A closer look

at the situational context of mobile communication

Veronika Karnowski a,, Olaf Jandura b,1

a Institut fr Kommunikationswissenschaft, und Medienforschung, LMU Mnchen, Oettingenstr., 67, 80538 Mnchen, Germanyb Kommunikations- und Medienwissenschaft, Heinrich-Heine-Universitt Dsseldorf, Universittsstr. 1, 40225 Dsseldorf, Germany

a r t i c l e i n f o

Article history:

Received 30 April 2012

Received in revised form 20 February 2013

Accepted 5 November 2013

Available online 11 November 2013

Keywords:

Mobile communication

Mobile web

Media use

Quantitative survey

Classification

Situational approach

a b s t r a c t

The web is going mobile, and the scope of mobile communication is widening tremen-

dously, thus paving the way for a wide array of new forms of mobile device use. However,

not every user is necessarily all the time taking advantage of the expanded affordances of

mobile devices. Texting and phoning are still the predominant services in mobile commu-

nication. Previous research has argued that different styles of mobile communication are

related to different user lifestyles. Thus, a remapping and matching of the landscape of

mobile communication in relation to user lifestyles seems necessary. In this paper, we take

one step back and first consider the instances in which lifestyles become behavior; i.e.

actual usage situations of mobile communication. We outline three empirically deduced

types of mobile communication usage situations, as well as three types of mobile web

usage situations, to shed light on the instant at which lifestyle becomes behavior; i.e. at

which specific usage situations of mobile communication actually occur.

2013 Elsevier Ltd. All rights reserved.

1. Introduction

The web is going mobilea trend which cannot be denied any longer. The scope of mobile communication has widenedtremendously, thus paving the way for a wide array of new forms of use. However, not every user is necessarily takingadvantage of these expanded possibilities of mobile phone use. Texting and phoning are still the predominant services inmobile communication (e.g.Boase and Ling, 2011). Previous researchers have argued that different styles of mobile commu-nication are related to the different lifestyles of users (e.g. Bouwmann et al., 2012). Thus a remapping and matching of thelandscape of mobile communication in relation to user lifestyles seems necessary.

In this paper, we take a step back and first consider instances in which lifestyle patterns become communication behav-iors; i.e. we examine actual usage patterns of mobile communication and their specific contexts. On the one hand, research in

the tradition of communication studies has been mostly blind to the situational contexts of new media usage; on the otherhand, research in information systems and computing has integrated context factors, but not precisely at the situational le-vel, and mostly from the viewpoint of technical artifacts rather than user standpoints. We would like to help close this gap byintegrating situational context factors in a user-centered analysis of mobile communication behavior. In this first step weoutline a classification of mobile communication usage situations based on situational contexts, and examine differencesamong the services used in relation to these situational contexts.

0736-5853/$ - see front matter 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.tele.2013.11.001

Corresponding author. Tel.: +49 89 2180 9495; fax: +49 89 2180 9429.

E-mail addresses:[email protected](V. Karnowski),[email protected](O. Jandura).1 Tel.: +49 211 81 10660; fax: +49 211 81 15212.

Telematics and Informatics 31 (2014) 184193

Contents lists available at ScienceDirect

Telematics and Informatics

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e/ t e l e
http://dx.doi.org/10.1016/j.tele.2013.11.001mailto:[email protected]:[email protected]://dx.doi.org/10.1016/j.tele.2013.11.001http://www.sciencedirect.com/science/journal/07365853http://www.elsevier.com/locate/telehttp://www.elsevier.com/locate/telehttp://www.sciencedirect.com/science/journal/07365853http://dx.doi.org/10.1016/j.tele.2013.11.001mailto:[email protected]:[email protected]://dx.doi.org/10.1016/j.tele.2013.11.001http://-/?-http://-/?-http://-/?-http://crossmark.crossref.org/dialog/?doi=10.1016/j.tele.2013.11.001&domain=pdf


2/10

2. Situational contexts and usage of new media services

Researchers commonly refer to the Theory of Planned Behavior (TPB;Ajzen, 1985) or various permutations thereof, toexplain the adoption and usage of new media services (e.g. see Bouwman et al., 2007; Wang et al., 2008).

The TPB (Ajzen, 1985) developed as an offshoot of the Theory of Reasoned Action (Fishbein and Ajzen, 1975), which takesinto account the influence of social norms on the adoption decision. According to the TPB, behavior is influenced not only byattitudes towards the behavior in question, but also by subjective norms and perceived behavioral control.

Attitudes towards a behavior consist of two interacting components: an individuals expectations regarding the conse-quences of the behavior in question, and his/her positive or negative evaluations of these consequences. Subjective normsrefer to the pressure exerted by the social surroundings of an individual, which influence the individual to execute or notto execute the behavior in question. Social norms also consist of two components: the individuals appraisal of what behavioris expected by his/her peers, and his/her evaluation of these expectations. Perceived behavioral control refers to the extent towhich an individual feels able to execute his/her behavior; it consists of both situational and internal dimensions. The sit-uational dimension describes the extent to which an individual objectively can execute a given behavior, while the internaldimension refers to whether the individual subjectively feels that he/she is able to execute the behavior (see Ajzen, 2005).

The most prominent extensions of TPB in the field of new media include the Technology Acceptance Model (TAM;Davis,1989), the Unified Theory of Acceptance and Use of Technology (UTAUT; Venkatesh et al., 2003), and the Mobile PhoneAppropriation model (MPA-model;Wirth et al., 2008). These theories elaborate on the TPB in terms of differentiating boththe factors that influence new media behaviors (TAM, UTAUT, MPA-model) and the actual forms of the new media behaviors(MPA-model). However, these theories have one prominent shortcoming: they do not include situational contexts. This lim-

itation becomes especially poignant when analyzing mobile communication. As mobile services are ubiquitous, the situa-tions that are involved, and their respective requirements, are virtually unlimited.Situational contexts of usage must, therefore, be integrated into our analysis. The information systems literature has dis-

cussed similar factors for over a decade, although they have generally been labeled as context factors. As early as Kristoffer-sen and Ljungberg (1999)pointed out that the use context of mobile handsets in work life varies widely across differentprofessions. Similarly,Perry et al. (2001) noted the influence of social and infrastructural factors on mobile computing. Insubsequent years, several studies in the field of information systems drew upon context factors when analyzing the adoptionand intended usage of various innovations, clearly stating the impact of these context factors on adoption and usage patterns(e.g.Mallat et al., 2009; Bouwman et al., 2012; Turner et al., 2008; van de Wijngaert and Bouwman, 2009). However, thesestudies did not focus on the usage situation per se, but were still operating under the assumptions of mixed and/or broaderlevels of analysis.

The analysis of single usage situations requires that we probe deeper, and ask what exactly constitutes a single usage sit-uation.Belk (1975)defined the environment of consumer behavior on the basis of five categories: physical, social, temporal,

task, and antecedent states. While not completely based on a specific situation, these categories provide us with insights intorelevant dimensions of situational contexts; i.e. physical and social. Similarly, Lee et al. (2005) and Vartiainen (2006)splitsituational contexts into physical and social (human) factors. The situational model ofZhang and Zhang (2012)follows sim-ilar lines, but focuses on media behaviors. Zhang and Zhangs model distinguishes between two interdependent factors influ-encing new media behaviors: personal psychologies on the one hand, and location-related conditions on the other (seeFig. 1). Personal psychologies are conceptualized alike gratifications sought, as in the Uses and Gratifications Approach(seeKatz et al., 1974; Palmgreen and Rayburn, 1985). Regarding location-related conditions, Zhang and Zhang differentiate

Fig. 1. The integrated model of computer multitasking (Zhang and Zhang, 2012, p. 1886).

V. Karnowski, O. Jandura / Telematics and Informatics 31 (2014) 184193 185
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


3/10

the dimensions of physical environments, media access, and social dynamics with respect to multitasking on and with mo-bile phones.

Location-related conditions include the physical environment, media access, and social dynamics. Physical environmentsconsist of a users familiarity with his/her surroundings; i.e. whether he/she is at home or in public, and whether he/she isfamiliar with his/her environment. Media access refers to whether the user has other opportunities for media access in addi-tion to a mobile phone, and if so, which? Media access is postulated to influence new media behaviors. Finally, the specificsocial dimensions in a situation form part of the location-related conditions. Is the user amongst peers or is he/she alone; i.e.who is in the users actual surroundings? Social dimensions also influence new media behavior.

By combining the TPB and its extensions with the Situational Theory of New Media Behaviors, we can conceptualize mo-bile web usage as influenced by user-related factors, location-related factors, and restrictions (see Fig. 2). User-related factorsrefer to the actual emotional state of a user, according to Bradley and Lang (1994). Location-related factors are differentiatedinto physical environments, media access, and social dynamics, according to Zhang and Zhang (2012; see above). Restrictionsare differentiated into financial, technical, temporal, and cognitive restrictions, according to Wirth et al. (2008).

Based on the theoretical framework outlined above, our research questions are as follows:

RQ1: In which situations, defined by physical environments, media access and social dynamics, do people use theirmobile devices?RQ2: What different types of situations can be distinguished?RQ3: Can we distinguish between a specific set of usage situations for mobile web usage as compared with other mobilecommunication services?

3. Method and measures

To answer the above research questions, we conducted an online survey in cooperation with Tomorrow Focus AG, one ofGermanys leading publicly traded internet companies. The survey was conducted during the period 25 November to 23December 2010, using a quantitative online questionnaire. To achieve our goal of collecting not a representative but a het-erogeneous sample, the questionnaire was linked to several websites which, in combination, reach two thirds of all Germaninternet users. The websites included news sites and special interest sites (e.g. pages for women, men, people interested infashion, etc.). However, because of privacy regulations, our survey results did not include the website that the survey par-ticipants had visited to take part in the survey.

The final sample consisted of 1400 individuals. The composition of the sample resembled that of average German dailyinternet users (ALLBUS, 2010): 72% of the respondents were male, 28% were female, the average age was 41 years, 55% re-ported a low to moderate degree of education, 36% reported a high degree of education, and 31% (432 respondents) browsed

the internet via a mobile device. Our data are consistent with the findings that in Germany, men tend to go online more oftenthan women, indicating that there is still a gender gap in internet usage in Germany ( Zillien, 2009).In terms of patterns of online usage with mobile devices, we found that: the mobile online group was a little younger on

average (average age, 37), and in this group the percentage of higher educated respondents was higher (43%), the percentageof respondents with a higher income was greater (15% vs. 9%, respectively), and the percentage of respondents with profes-sional lifestyles was greater (84% vs. 72%, respectively). Because of the substantial similarity of the sample to representativesamples of German daily online users, we refrained from weighing the sample using sociodemographic variables and thefrequency of the internet use.

Fig. 2. Factors influencing mobile web usage.

186 V. Karnowski, O. Jandura / Telematics and Informatics 31 (2014) 184193
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


4/10

The questionnaire contained questions that provided us with information on four online usage dimensions. The firstdimension focused on the prerequisites of mobile internet usage. Respondents were asked whether or not they possesseda mobile device capable of online access; if so, the respondents were asked if they actually used the device to access theinternet, and if so, when was the last time that they had used the device in this way.

The second dimension was used to characterize the last usage episode. Because mobile internet use was in an early stage atthe time of the survey, we asked about the last time that the respondent had actually gone online via a mobile device. In thisway, we gained information only on usage situations relevant to our study. By initially addressing users with a sudden pop-up window during actual internet sessions, we were assured that:

(1) The times of day that online users were asked to take part in the survey varied, as did the last mobile internet usagesituation; 29% of the questionnaires were completed in the morning (6 amnoon), 15% in the afternoon (noon5 pm),16% in the early evening (5 pm8 pm), 21% in the late evening (8 pmmidnight), and 21% at night (midnight6 am).

(2) Participants took part in the survey only once, thus maintaining the statistical independence between personal andsituational variables.

(3) The questionnaire was not completed only on weekends, as internet usage patterns on weekends typically differ fromthose on weekdays.

Because our goal was to analyze the last usage episode in detail, we asked about aspects related to: the online sites thatwere visited, the respondents location (nominal scale: at home/at work/en route), the presence of other people (nom-inal scale: family/friends/colleagues/strangers/nobody), whether the location they were in when they accessed theinternet was familiar (nominal scale: yes/no), and whether alternative media services were available at the time (news-papers, magazines, TV, radio, internet via desktop/laptop). We also asked about the respondents mood during online access,measured by the self-assessment manikin (SAM), a non-verbal pictorial assessment technique (Bradley and Lang, 1994). Fi-nally, we asked about possible restrictions on access (temporal, cognitive, technical, financial; Wirth et al., 2008). Eachdimension was explored using two items, and respondents were asked about the extent to which they agreed or disagreedwith each item, using a 7-point Likert scale.

Third, we measuredmotives and gratifications. In line with previous studies, we asked respondents about their motivesand gratification in relation to their online experiences using a mobile device. We included the following dimensions accord-ing to the previous studies ofLeung and Wei (2000), LaRose and Eastin (2004) and Wei (2008): information, entertainment,maintaining relationships, status, and availability/access. A total of 21 gratification items were measured. Each dimensionwas explored using four to five items. A 5-point Likert scale was used to determine the extent to which respondents agreedor disagreed with each item.

3.1. Control measures

We measured demographic variables such as age, gender, region (nominal scale: village/small city/big city), income(3-point scale: low/middle/high), formal education (3-point scale: low/middle/high), and currently employed(nominal scale: yes/no). In addition, we asked how often, typically, the respondent was in an unknown location (5-pointLikert scale, ranging from very often to not at all) and how often he/she spends time traveling (5-point Likert scale, rang-ing from very often to not at all).

4. Results

4.1. Mobile communication in general

To identify different classes of mobile communication usage situations, we used the clustering technique of latent classanalysis (LCA). Latent class analysis has a variety of advantages over traditional cluster analysis, as it allows for the classi-fication of variables at each level of measurement, and different levels of measurement can be integrated into the analysis. Incontrast to traditional cluster analysis, LCA does not necessarily result in a cluster solution, and it can also reject clustering ofthe data (see Fraley and Raftery, 1998). Latent class analysis provides statistical tests to identify the exact number of clusters.Accordingly, it is less arbitrary than traditional cluster analysis. On account of its probabilistic conception, LCA also takes intoaccount the possibility that the clustered variables might not be completely reliable or completely valid.

As outlined above, usage situations consist of user-related and location-related aspects, as well as restrictions. Unfortu-nately, the variance of the emotional state of our respondents, as well as that of perceived restrictions, was low throughoutour sample. Thus, only location-related aspects of mobile web usage situations, as well as the availability of data services onthe mobile device in a specific situation, could be integrated into the LCA.

To identify the number of clusters, we first calculated and compared one- to ten-cluster solutions. All but the one-clustersolution show a non-significantp-value of the likelihood-ratio test and the CressieRead test; thus, the model predictions didnot differ significantly from the observed data. A check of the likelihood-ratio test by both the bootstrapping method andPearsons v2 also indicates a significant p-value for the two-cluster solution, which can thus be eliminated.

http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


5/10

Consequently only the three-, four-, and six- to ten-cluster solutions were retained in the analysis (see Table 1). Generally,the solution including the fewest parameters to be estimated, and thus the lowest BIC value, is considered the most suitable.The three-cluster solution meets this criterion.

Based on the LCA, there are specific probabilities of the different parameter values integrated into the analysis for eachcluster. The different clusters can be described on the basis of these probabilities (see Table 2).

Cluster 1: Mobile@home

In the evening, Sarah is watching TV at home. All of a sudden, she remembers that it is her cousins birthday today. As it is al-ready too late to call, she quickly sends a text to her cousin saying, Happy Birthday!.

This is by far the largest cluster, containing 56.5% of all usage situations. These usage situations most likely occur in afixed and highly familiar location (such as at home).

Cluster 2: En route

Mr. Smith is on his way to meet a former class mate. Unfortunately, he left the exact address at home. So, he tries to reach some

of his other former class mates to ask for the address. As he does not have a flat rate plan, he wants to keep the conversation as short

as possible.Usage situations in this cluster represent mobile communications occurring while the user is on his/her way. The social

surroundings are mostly unknown to the user, and the familiarity with the location varies widely. The availability of onlineservices is least probable in this cluster. This cluster is far smaller than cluster 1 (including only 23.8% of all usage situations).

Cluster 3: Hanging out with peers

Tom and some of his mates are sitting in a bar. They all have flat rate plans for their cell phones. Throughout the evening, they

are constantly posting photographs and comments on Facebook.This usage cluster, which includes 19.8% of all usage situations, is the smallest usage cluster. Here, usage most probably

occurs en route, but could also be in a fixed location. The social surroundings are well known to the user, but the familiarity

Table 1

Likelihood-ratio test (including bootstrapping), CressieRead test, Pearsons v2, and the Bayesian information criterion (BIC) for the one- to ten-cluster

solutions for mobile communication usage situations.

Table 2

Average probabilities of parameter values, explained variance of classified variables, and relative cluster size of the three-cluster solution for the mobile

communication usage situations.

Cluster 1 (%) Cluster 2 (%) Cluster 3 (%) R2 (%)

Online services available on mobile device 67 56 75 1.9

Fixed location 100 0 17 87.7

On the road 0 99 99 97.4

Known social surroundings 53 5 78 25.2

Financial restrictions

low 56 52 61 0.5

middle 33 34 30

high 12 14 9

Familiarity with location

low 2 17 20 15.4

middle 13 30 31

high 85 53 49

Relative size 56.5 23.8 19.8



6/10

with the location is low. The probability of the availability of online services is higher in this cluster than in any of the otherclusters.

We now compare the classes of mobile communication usage situations based on sociodemographic parameters, grati-fications sought, and services used. To do so, each case is attributed to the cluster it most probably belongs to. The classifi-cation error in this procedure (i.e. the percentage of cases which are classified incorrectly) is 5.1%.

Despite the overall predominance of males in our sample, the usage cluster Hanging out with peers occurs more often formale than for female users (p< 0.05); the same is true for users with a lower education level; that is, Hanging out with peers

occurs more often amongst those with low than those with medium or high education levels (p< 0.05). In addition, users inthe usage cluster En route are significantly older than are all other users in our sample (seeTable 3).

Overall, telephony is the dominant usage of mobile phones. This service is used significantly more often in En route sit-uations than in Mobile@home and Hanging out with peers situations. Text messaging occurs most often in Mobile@homesituations. Other services are most often used when Hanging out with peers. For the other mobile phone services, there areno significant differences in usage between the three types of mobile communication situations (see Table 4).

Items pertaining to gratification of mobile communication were subsumed to five indices: status, maintaining relation-ships, entertainment, access, and information. According to Cronbachs alpha values, the internal reliability of these indices issatisfactory (seeTable 5).

Table 3

Sociodemographic parameters by mobile communication usage situations.

Mobile@home(n= 4 83) (%) En route(n= 381) (%) Hanging out with peers(n= 87) (%) F-Value g2 (%)

Sex

Male 70a 74a 81b 3.8 0.6

Female 30a 26a 18b

Educational level

Low 18a 15a 25b 3.0 0.5

Medium 25 25 19 1.1 0.2

High 44 50 44 3.0 0.5

Age 38.9a 44.8b 38.9a 20.6 3.1

p< 0.05, p< 0.01, p < 0.001; means marked by different characters differ significantly.

Table 4

Percentage of services used by mobile communication usage situations.

Service used Mobile@home (n= 483)

(%)

En route (n= 381)

(%)

Hanging out with peers (n= 87)

(%)

F-

Value

g2

(%)

Telephony 45a 59b 43a 13.7 2.1

SMS/MMS 27a 14b 17b 16.1 2.5

Radio/MP3 4 4 3 0.2 0.0

E-mail 5 5 5 0.1 0.0

Internet via browser 6 7 12 2.6 0.4

Internet via app 3 4 4 0.5 0.1

Games 3 2 2 0.5 0.1

Other (camera, calendar, alarm clock,

. . .)

9a 6a 15b 6.2 1.0

p< 0.05,

p< 0.01,

p < 0.001; means marked by different characters differ significantly.

Table 5

Means and Cronbachs alpha values for gratification indices of the mobile communication usage situations.

Gratification indices Number of

items

Cronbachs

alpha

Mobile@home

(n= 483)

En route

(n= 381)

Hanging out with peers

(n= 87)

F-

Value

Status 4 0.74 4.5a 4.7a 4.0b 10.1

Maintaining

relationships

5 0.77 3.6a 3.7a 2.7b 11.5

Entertainment 4 0.87 3.0 3.3 3.0 2.1

Access 4 0.76 3.6a 3.7a 3.0b 5.5

Information 4 0.70 3.0a 3.0a 2.2b 5.9

Agreement on a 5-point scale, from 1 = strongly agree to 5 = strongly disagree. p< 0.05, p< 0.01, p< 0.001, means marked by different characters differ

significantly.

http://-/?-http://-/?-http://-/?-http://-/?-


7/10

Significant differences occur in all but one dimension of gratifications sought between the clusters Mobile@home and Enroute and the cluster Hanging out with peers. In situations of Hanging out with peers, users assign higher importance tostatus, maintaining relationships, access, and information, than do users in other usage situations.

4.2. Mobile web use

In this section, we concentrate on mobile web use, which is a specific sub-set of mobile communications. In our sample,

432 respondents were mobile online users, and each of these respondents reported their last mobile web usage situation. Toidentify specific classes of mobile web usage situations, we used the LCA clustering technique.

To identify the number of clusters, we first calculated and compared the one- to ten-cluster solutions. All but the one-cluster solution show non-significantp-values for the likelihood-ratio test and the CressieRead test; the same results wereobtained using the bootstrapping method. Pearsons v2, on the other hand, indicates significant p-values for the two-clusterand five-cluster solutions, which can thus be eliminated. As the three-cluster solution has the lowest BIC value, this solutionis the most suitable (seeTable 6).

We now describe these three clusters based on the specific probabilities of the different parameter-values integrated inthe analysis for each cluster (seeTable 7).

Cluster 1: On the way

Each morning, Mr. B is sitting in the metro on his way to work. Sometimes his neighbor is sitting beside him, as they are both

commuting. Some years ago, he used to buy a newspaper to read on the train, but he does not do so any longer. Each morning whilecommuting, he is checking his favorite news sites via his smartphone.

Table 6

Likelihood-ratio (incl. Bootstrapping), CressieRead, Pearsons v2, and Bayesian information criterion (BIC) tests of the one- to ten-cluster solutions for mobile

web usage situations.

Table 7

Average probabilities of parameter values, explained variance of classified variables, and relative cluster size of the three-cluster solution for mobile web usage

situations.

http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


8/10

On the way is the largest cluster of mobile web usage situations. Usage situations in this cluster occur while the user is intransit. In these circumstances, the user has no, or nearly no, competing media access to mobile web usage. As comparedwith the circumstances in the other two clusters, the social surroundings are rather unfamiliar to the user, and the proba-bility of a rather unfamiliar location is highest in this cluster.

Cluster 2: Homezone

Heather is at home watching a new serial on TV. During the commercial break, she pulls out her smartphone and browses

through her friends Facebook posts. She could check them using her PC, but she is too lazy to get up, go to her desk, and startthe PC.These usage situations most probably occur in private and very familiar locations. Possibilities of alternative media access

are rather high. The probability that known individuals are present in the users social surroundings is average; i.e. the usermight be in the companionship of others, but he/she might also be alone.

Cluster 3: Work or friends

Mark and Kate are spending an evening at their favorite bar together with their friends. Mark tells them about a book he is

currently reading, but he can not remember the authors name. Thus, he takes out his smartphone and Googles the authors details .The probability of being together with peers or family is higher in this cluster than in the other clusters. The usage sit-

uations occur within fixed and quite familiar locations, which are not necessarily in the private realm. Additional media ac-cess varies within these usage situations, but might be rather low.

We compared the usage clusters using the exogenous variables of age, gender, service used, and gratifications sought. To

do so, each case was attributed to the cluster it most probably belongs to, which resulted in a classification error of 8.1%. Nodifferences were detected among the different types of mobile web usage situations with respect to users gender or educa-tional level. Users in the cluster On the way were significantly older than other mobile online users (see Table 8).

News and looking up information are the most frequent activities when using an online service with the mobile phone.Looking up information occurs most often in Work or friends situations, next most often in On the way situations, and leastoften in (but still in 29% of) situations in which the user is in his/her Homezone. Significant differences occur with respect tomobile use of TV and video platforms, which occur significantly more often when with Work or friends than On the way orin the Homezone. The same is true for the mobile use of web radio (seeTable 9).

Significant differences occur between the three clusters in terms of the three gratification dimensions: maintaining rela-tionships, entertainment, and information. Maintaining relationships is of lower relevance when using mobile web servicesin situations of Work and Friends, as compared with the other two situations. Because users are likely already in the com-panionship of peers in these usage situations, additional social contact via the mobile web is of no importance in suchsituations.

Table 8

Sociodemographics by mobile web usage situations.

On the way (n= 1 93) (%) Homezone (n= 174) (%) Work or friends (n= 65) (%) F-Value g2 (%)

Sex

Male 80 83 85 0.4 0.2

Female 20 17 15

Educational level

Low 12 11 12 0.0 0.0

Medium 25 24 23 0.1 0.1

High 56 50 51 0.8 0.4

Age 39.1a 35.1b 34.9b 4.1 1.9

p< 0.05, p< 0.01, p < 0.001, means marked by different characters differ significantly.

Table 9

Percentage of services used by mobile web usage situations.

Service used On the way (n= 193) (%) Homezone (n= 174) (%) Work or friends (n= 65) (%) F-Value g2 (%)

News 32 32 37 0.4 0.2

E-commerce 7 6 15 2.8 1.3

Looking up information 42a 29b 55c 7.8 3.5

Social networks 20 27 32 2.3 1.1

Online gaming 3 3 5 0.3 0.2

Chat/instant messaging 12 10 15 0.7 0.3

TV or video platforms 4a 9a 18b 7.7 3.5

Web radio 4a 4a 12b 3.7 1.7

Other 2 3 0 1.2 0.5

p< 0.05, p< 0.01, p < 0.001, means marked by different characters differ significantly.



9/10

On the way and Homezone are significantly different with respect to entertainment. Entertainment is more relevantwhen On the way than when at home, where one has access to a variety of other distractions. Information is also of leastrelevance in usage situations involving Work or friends; i.e. when in the presence of others, but most probably deprived ofother media access (seeTable 10).

5. Discussion, limitations and conclusion

The Situational Theory of New Media Behaviors provides a promising approach for the analysis of mobile communicationsituations generally, and of mobile communication for online usage specifically. Our online survey identified three mainsituations in which mobile communication occurs: the first is when users are among known peers and in familiar locations,a situation which occurs mostly at home (Mobile@home); the second is when users are on the way and in unknownsurroundings and among unknown people (En route); and the third is when users are with peers but in unknown locations,such as in a restaurant or bar (Hanging out with peers). The occurrence of these mobile communication usage situationsvaries according to age, gender, and educational level. In addition, services used and gratifications sought are differentamong the different usage clusters.

A closer look at mobile internet use reveals a subset of similar mobile internet usage patterns. Not surprisingly, mobilephones are used to log onto the internet when people are in unfamiliar locations and in public situations (On the way). Justas predictable is the finding that respondents used their mobiles to go online when they were in fixed and familiar locations,such as in bars or plazas, when access to other helpful media were limited or unavailable (Work or friends). A rather unex-pected finding, however, was the occurrence of many situations in which users, despite opportunities to use conventionalPCs or laptops, went online using their mobile devices while at home (Homezone). This find underscores the importanceof mobile communication via mobile devices in general, and internet usage in particular, in fixed locations. In terms of soci-odemographic factors, the variance among these clusters is very low; however, the clusters do differ in terms of mobile webservices used; i.e. entertainment-related aspects are of greater importance when at work or with friends (Work or friends),whereas information and news services are most frequently accessed when On the way or in the Homezone.

Although we successfully identified three main mobile communication usage situations, as well as three main mobileinternet usage situations, we feel that further research is required in this area on account of the theoretical and methodo-logical constraints of our study. First, the question arises as to whether the high percentage of situations in which people usecell phones at home is an artifact produced by the questionnaire. The respondents encountered our survey and were askedabout their last mobile communication use in a situation in which they were already accessing the internet. Often, peoplewill have done so in a fixed location, which, in the majority of cases, is probably in the so-called Homezone. Thus, we can-not rule out the possibility that our study overestimated the number of mobile usage situations at home.

Second, the measurements of the emotional state of the participants can be questioned. It appears that, despite the use ofa well-tested scale, we ended up with rather nonspecific data because respondents were required to assess their moods ret-rospectively. We assume that our participants were unable to truly remember how they felt during their last mobile onlineusage situation. If we are correct about this shortcoming, then our method of mood measurement during online sessions isquestionable; i.e. an in situ measurement such as the experience sampling method might be required (see Hektner et al.,2007; Karnowski and Doedens, 2010; Larson and Csikszentmihalyi, 1983).

Third, we suggest another reason for the fact that we obtained rather nonspecific responses to the questions dealing withtemporal, cognitive, technical, and financial restrictions on mobile communication. On account of technical improvements inmobile devices, it is possible that Wirth et al.s (2008) questionnaire items, which were developed four years before the pres-ent study was implemented, were already outdated at the time of the study, and that the relevant restrictions cannot bemeasured validly at this time.

Fourth, our data characterize the situation in a western European nation prior to the widespread use of mobile internetdevices, when online usage was still an activity that was very much linked to traditional desktops and laptops.

Finally, we must bear in mind that our survey is cross-sectional. Therefore, we were only able to gather information onone single usage situation. Because of the one-shot nature of the survey, and our decision to deal with actual online sessionsrather than self-assessment summaries of individuals mobile internet use, we were only able to collect data for one situa-tion; i.e. the two concepts are confounded. We therefore recommend that future research produce longitudinal data for theanalysis of mobile usage patterns.

Table 10

Gratification indices by mobile web usage clusters.

On the way (n= 192) Homezone (n= 170) Work or friends (n= 64) F-Value g2 (%)

Status 4.3 4.2 4.0 2.40 1.2

Maintaining relationships 3.5a 3.3a 2.9b 7.27 3.3

Entertainment 3.2a 2.8b 3.0ab 5.47 2.5

Access 3.5 3.3 3.1 2.19 1.1

Information 2.7a 2.7a 2.3b 4.51 2.1

Agreement on a 5-point scale, from 1 = strongly agree to 5 = strongly disagree. p< 0.05, p< 0.01, p< 0.001, means marked by different characters differ

significantly.



10/10

References

Ajzen, I., 1985. From intentions to actions: a theory of planned behavior. In: Kuhl, J., Beckman, J. (Eds.), Action-control: From Cognition to Behavior. Springer,Heidelberg, pp. 1139.

Ajzen, I., 2005. Attitudes, Personality and Behavior, second ed. Open University Press, New York.Belk, R.W., 1975. Situational variables and consumer behavior. Journal of Consumer Research 2, 157164 .Boase, J., Ling, R., 2011. Measuring mobile phone use: self-report versus log data. ICA Pre-Conference 2011 Seamlessly Mobile? Mobile Communication @ a

Crossroads, 25./26.5.2011, Boston, USA.Bouwman, H., Bejar, A., Nikou, S., 2012. Mobile services put in context: a Q-sort analysis. Telematics and Informatics 29, 6681.

Bouwman, H., Carlsson, C., Molina-Castillo, F.J., Walden, P., 2007. Barriers and drivers in the adoption of current and future mobile services in Finland.Telematics and Informatics 24, 145160.Bouwmann, H., Lopz-Nicols, C., Molina-Castillo, F.-J., van Hattum, P., 2012. Consumer lifestyles: alternative adoption patterns for advanced mobile

services. International Journal of Mobile Communications 10, 169189.Bradley, M.M., Lang, P.J., 1994. Measuring emotion: the self-assessment manikin and the semantic differential. Journal of Behavioral Therapy and

Experimental Psychiatry 25, 4959.Davis, F.D., 1989. Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly 13, 319340 .Fishbein, M., Ajzen, I., 1975. Believe, Attitude, Intention and Behavior: An Introduction to Theory and Research. Addison-Wesley, Reading.Fraley, C., Raftery, A., 1998. How many clusters? Which clustering method? Answers via model-based cluster analysis. The Computer Journal 41, 578588 .Hektner, J.M., Schmidt, J.A., Csikszentmihalyi, M., 2007. Experience sampling method: measuring the quality of everyday life. Sage, Thousand Oaks .Karnowski, V., Doedens, S., 2010. Mobile experience sampling. Eine Methode zur Untersuchung mobilen Mediennutzungsverhaltens [Mobile Experience

Sampling. A method for analyzing mobile media usage]. In: Jackob, N., Zerback, T., Jandura, O., Maurer, M. (Eds.), Methoden der Online-Forschung: DasInternet als Forschungsinstrument und -gegenstand der Kommunikationswissenschaft. Verlag Herbert von Halem, Kln, pp. 211226.

Lee, I., Kim, J., Kim, J., 2005. Use contexts for the mobile internet: a longitudinal study monitoring actual use of mobile internet services. InternationalJournal of HumanComputer Interaction 18, 269292.

Katz, E., Blumler, J.G., Gurevitch, M., 1974. Utilization of mass communication by the individual. In: Blumler, J.G., Katz, E. (Eds.), The Uses of MassCommunications. Current Perspectives in Gratifications Research. Sage, Beverly Hills, pp. 168249.

Kristoffersen, S., Ljungberg, F., 1999. Making Place to Make IT Work: Empirical Explorations of HCI for Mobile CSCW. GROUP 99 Proceedings of theinternational ACM SIGGROUP conference on Supporting group work, 276285.LaRose, R., Eastin, M., 2004. A social cognitive theory of internet uses and gratifications: toward a newmodel of media attendance. Journal of Broadcasting &

Electronic Media 48, 358377.Larson, R., Csikszentmihalyi, M., 1983. The experience sampling method. In: Reis, H.T. (Ed.), Paperback Sourcebooks in the Jossey-Bass Social and Behavioral

Science Series. Naturalistic Approaches to Studying Social Interaction. Jossey-Bass, San Francisco, pp. 4156.Leung, L., Wei, R., 2000. More than just talk on the move: uses & gratifications of the cellular phone. Journalism and Mass Communication Quarterly 77,

308320.Mallat, N., Rossi, M., Tuunainen, V.K., rni, A., 2009. The impact of use context on mobile services acceptance: the case of mobile ticketing. Information &

Management 45, 190195.Palmgreen, P., Rayburn, J.D., 1985. An expectancy-value approach to media gratifications. In: Rosengren, K.E., Wenner, A., Palmgreen, P. (Eds.), Media

Gratifications Research. Current Perspectives. Sage, Beverly Hills, pp. 6172.Perry, M., Ohara, K., Sellen, A., Brown, B., Harper, R., 2001. Dealing with mobility: understanding access anytime, anywhere. ACM Transactions on

ComputerHuman Interaction 8, 323347.Turner, M., Love, S., Howell, M., 2008. Understanding emotions experienced when using a mobile phone in public: the social usability of mobile (cellular)

telephones. Telematics and Informatics 25, 201215.van de Wijngaert, L., Bouwman, H., 2009. Would you share? Predicting the potential use of a new technology. Telematics and Informatics 26, 85102 .Vartiainen, M., 2006. Mobile virtual work concepts, outcomes and challenges. In: Andriessen, E.J.H., Vartiainen, M. (Eds.), Mobile Virtual Work. A new

paradigm? Springer, Berlin, pp. 1344.Venkatesh, V., Morris, M.G., Davis, G.B., Davis, F., 2003. User acceptance of information technology: toward a unified model. MIS Quarterly 27, 425478.Wang, C.C., Lo, S.K., Fang, W., 2008. Extending the technology acceptance model to mobile telecommunication innovation: the existence of network

externalities. Journal of Consumer Behaviour 7, 101110.Wei, R., 2008. Motivations for using the mobile phone for mass communications and entertainment. Telematics and Informatics 25, 3646 .Wirth, W., Von Pape, T., Karnowski, V., 2008. An integrative model of mobile phone appropriation. Journal of Computer-mediated Communication 13, 593

617.Zhang, W., Zhang, L., 2012. Explicating multitasking with computers: gratifications and situations. Computers in Human Behavior 28, 18831891 .Zillien, N., 2009. Digitale Ungleichheit: Neue Technologien und alte Ungleichheiten in der Informations- und Wissensgesellschaft [Digital inequality: new

technologies and old inequalities in an information and knowledge society]. Vs Verlag, Wiesbaden .

http://refhub.elsevier.com/S0736-5853(13)00068-3/h0005http://refhub.elsevier.com/S0736-5853(13)00068-3/h0005http://refhub.elsevier.com/S0736-5853(13)00068-3/h0010http://refhub.elsevier.com/S0736-5853(13)00068-3/h0015http://refhub.elsevier.com/S0736-5853(13)00068-3/h0025http://refhub.elsevier.com/S0736-5853(13)00068-3/h0030http://refhub.elsevier.com/S0736-5853(13)00068-3/h0030http://refhub.elsevier.com/S0736-5853(13)00068-3/h0035http://refhub.elsevier.com/S0736-5853(13)00068-3/h0035http://refhub.elsevier.com/S0736-5853(13)00068-3/h0035http://refhub.elsevier.com/S0736-5853(13)00068-3/h0040http://refhub.elsevier.com/S0736-5853(13)00068-3/h0040http://refhub.elsevier.com/S0736-5853(13)00068-3/h0045http://refhub.elsevier.com/S0736-5853(13)00068-3/h0050http://refhub.elsevier.com/S0736-5853(13)00068-3/h0050http://refhub.elsevier.com/S0736-5853(13)00068-3/h0055http://refhub.elsevier.com/S0736-5853(13)00068-3/h0060http://refhub.elsevier.com/S0736-5853(13)00068-3/h0065http://refhub.elsevier.com/S0736-5853(13)00068-3/h0065http://refhub.elsevier.com/S0736-5853(13)00068-3/h0065http://refhub.elsevier.com/S0736-5853(13)00068-3/h0070http://refhub.elsevier.com/S0736-5853(13)00068-3/h0070http://refhub.elsevier.com/S0736-5853(13)00068-3/h0070http://refhub.elsevier.com/S0736-5853(13)00068-3/h0080http://refhub.elsevier.com/S0736-5853(13)00068-3/h0080http://refhub.elsevier.com/S0736-5853(13)00068-3/h0090http://refhub.elsevier.com/S0736-5853(13)00068-3/h0090http://refhub.elsevier.com/S0736-5853(13)00068-3/h0095http://refhub.elsevier.com/S0736-5853(13)00068-3/h0095http://refhub.elsevier.com/S0736-5853(13)00068-3/h0100http://refhub.elsevier.com/S0736-5853(13)00068-3/h0100http://refhub.elsevier.com/S0736-5853(13)00068-3/h0105http://refhub.elsevier.com/S0736-5853(13)00068-3/h0105http://refhub.elsevier.com/S0736-5853(13)00068-3/h0110http://refhub.elsevier.com/S0736-5853(13)00068-3/h0110http://refhub.elsevier.com/S0736-5853(13)00068-3/h0115http://refhub.elsevier.com/S0736-5853(13)00068-3/h0115http://refhub.elsevier.com/S0736-5853(13)00068-3/h0125http://refhub.elsevier.com/S0736-5853(13)00068-3/h0125http://refhub.elsevier.com/S0736-5853(13)00068-3/h0125http://refhub.elsevier.com/S0736-5853(13)00068-3/h0130http://refhub.elsevier.com/S0736-5853(13)00068-3/h0135http://refhub.elsevier.com/S0736-5853(13)00068-3/h0135http://refhub.elsevier.com/S0736-5853(13)00068-3/h0140http://refhub.elsevier.com/S0736-5853(13)00068-3/h0145http://refhub.elsevier.com/S0736-5853(13)00068-3/h0145http://refhub.elsevier.com/S0736-5853(13)00068-3/h0150http://refhub.elsevier.com/S0736-5853(13)00068-3/h0155http://refhub.elsevier.com/S0736-5853(13)00068-3/h0155http://refhub.elsevier.com/S0736-5853(13)00068-3/h0160http://refhub.elsevier.com/S0736-5853(13)00068-3/h0160http://refhub.elsevier.com/S0736-5853(13)00068-3/h0165http://refhub.elsevier.com/S0736-5853(13)00068-3/h0165http://refhub.elsevier.com/S0736-5853(13)00068-3/h0165http://refhub.elsevier.com/S0736-5853(13)00068-3/h0165http://refhub.elsevier.com/S0736-5853(13)00068-3/h0160http://refhub.elsevier.com/S0736-5853(13)00068-3/h0155http://refhub.elsevier.com/S0736-5853(13)00068-3/h0155http://refhub.elsevier.com/S0736-5853(13)00068-3/h0150http://refhub.elsevier.com/S0736-5853(13)00068-3/h0145http://refhub.elsevier.com/S0736-5853(13)00068-3/h0145http://refhub.elsevier.com/S0736-5853(13)00068-3/h0140http://refhub.elsevier.com/S0736-5853(13)00068-3/h0135http://refhub.elsevier.com/S0736-5853(13)00068-3/h0135http://refhub.elsevier.com/S0736-5853(13)00068-3/h0130http://refhub.elsevier.com/S0736-5853(13)00068-3/h0125http://refhub.elsevier.com/S0736-5853(13)00068-3/h0125http://refhub.elsevier.com/S0736-5853(13)00068-3/h0115http://refhub.elsevier.com/S0736-5853(13)00068-3/h0115http://refhub.elsevier.com/S0736-5853(13)00068-3/h0110http://refhub.elsevier.com/S0736-5853(13)00068-3/h0110http://refhub.elsevier.com/S0736-5853(13)00068-3/h0105http://refhub.elsevier.com/S0736-5853(13)00068-3/h0105http://refhub.elsevier.com/S0736-5853(13)00068-3/h0100http://refhub.elsevier.com/S0736-5853(13)00068-3/h0100http://refhub.elsevier.com/S0736-5853(13)00068-3/h0095http://refhub.elsevier.com/S0736-5853(13)00068-3/h0095http://refhub.elsevier.com/S0736-5853(13)00068-3/h0090http://refhub.elsevier.com/S0736-5853(13)00068-3/h0090http://refhub.elsevier.com/S0736-5853(13)00068-3/h0080http://refhub.elsevier.com/S0736-5853(13)00068-3/h0080http://refhub.elsevier.com/S0736-5853(13)00068-3/h0070http://refhub.elsevier.com/S0736-5853(13)00068-3/h0070http://refhub.elsevier.com/S0736-5853(13)00068-3/h0065http://refhub.elsevier.com/S0736-5853(13)00068-3/h0065http://refhub.elsevier.com/S0736-5853(13)00068-3/h0065http://refhub.elsevier.com/S0736-5853(13)00068-3/h0060http://refhub.elsevier.com/S0736-5853(13)00068-3/h0055http://refhub.elsevier.com/S0736-5853(13)00068-3/h0050http://refhub.elsevier.com/S0736-5853(13)00068-3/h0045http://refhub.elsevier.com/S0736-5853(13)00068-3/h0040http://refhub.elsevier.com/S0736-5853(13)00068-3/h0040http://refhub.elsevier.com/S0736-5853(13)00068-3/h0035http://refhub.elsevier.com/S0736-5853(13)00068-3/h0035http://refhub.elsevier.com/S0736-5853(13)00068-3/h0030http://refhub.elsevier.com/S0736-5853(13)00068-3/h0030http://refhub.elsevier.com/S0736-5853(13)00068-3/h0025http://refhub.elsevier.com/S0736-5853(13)00068-3/h0015http://refhub.elsevier.com/S0736-5853(13)00068-3/h0010http://refhub.elsevier.com/S0736-5853(13)00068-3/h0005http://refhub.elsevier.com/S0736-5853(13)00068-3/h0005

veronika karnowski a,⇑

Documents