Content uploaded by Thomas Hess
Author content
All content in this area was uploaded by Thomas Hess on Dec 19, 2018
Content may be subject to copyright.
Oechslein et al. Paying for News: Opportunities through PNAs
Proceedings of the Nineteenth Americas Conference on Information Systems, Chicago, Illinois, August 15-17, 2013. 1
Paying for News: Opportunities for a New Business Model
through Personalized News Aggregators (PNAs)
Completed Research Paper
Oliver Oechslein
Ludwig-Maximilians-Universität München
oechslein@bwl.lmu.de
Thomas Hess
Ludwig-Maximilians-Universität München
thess@bwl.lmu.de
ABSTRACT
News consumption has been evolving from offline newspapers to online news. However, while offline newspapers sales are
decreasing, online news business models have never been entrenched. Meanwhile, the new technology of social
recommender systems enable automated news aggregation. Personalized news aggregators (PNAs) rely on this technology,
and provide personalized news in visually appealing ways that might deliver the potential for a new business model.
However, there is no research on PNA configuration or users’ willingness to pay (WTP).
An empirical investigation with 116 participants examined usage features influencing PNA users’ adoption and their WTP
for a paid-based service. First, we showed that perceived usefulness, usage comfort, awareness, and (social) personalization
significantly influence intention to use a PNA. Users are also considering price. Second, we found an optimal price point of
1.88€ and a price range up to 6.83€ for monthly use.
Keywords
Personalized news aggregator, business model, social recommender system.
INTRODUCTION
Traditionally, news has been provided by newspapers, and the business model of selling newspapers or advertisements has
been around for some time. Nevertheless, due to digitalization, newspapers’ sales figures have dropped and the traditional
business model might no longer fit anymore. The transformation of an offline to an online business model did work, but in
the long run the consumer has many other options through which to consume news complementary. Besides, consumers’
WTP for online content is low (Dou, 2004). To date, publishers still have problems to find an appropriate digitalization
strategy in order to monetize news and content and to counteract the decline in revenue.
Online news has an important advantage for consumers: it is possible to adjust news according to preferences. In the
literature, it turned out that there is a correlation between online news, personalization, and new potential business revenue
strategies (Saeaeksjaervi, Wagner and Santonen, 2003). Meanwhile, the new technology of social recommender systems have
improved content personalization and adaptation to a user’s preferences. Content bundling can be transformed from manual
bundling to an automated aggregation. This technology is being used in new types of services: personalized news aggregators
(PNAs). The service is mostly optimized for mobile devices (e.g. tablet computers) and presents a personalized selection of
news and other content sources in an optically unified interface. A first approach to implementing this service is Flipboard.
There is little research about PNAs in the information systems (IS) literature available (e.g., Nanas, Vavalis and Houstis,
2010), and none on PNA configuration in a media context. A service such as PNA might enable the establishment a new
digital business model for news. This new content distribution form has not yet been explored and requires, besides the
investigation of the underlying technology, a scientific research of economic effects. To design a profitable business model
and digitalization strategy for publishers, it is necessary to know which features are in fact needed for the service and to know
what the user is willing to pay for it. To provide a first insight, this study examines the importance of different features in
order to influence users’ intention to use a PNA. We also examine WTP for this service, to show if there is a general
opportunity for a paid-based business model. We address the following research questions:
RQ1: Which features are relevant from a user perspective and how do they influence the intention to use a PNA?
RQ2: What is user WTP for PNA service?
Oechslein et al. Paying for News: Opportunities through PNAs
Proceedings of the Nineteenth Americas Conference on Information Systems, Chicago, Illinois, August 15-17, 2013. 2
The structure of the paper is as follows: First, we present a review of technologies and business models. We then present the
development of our research model, the hypotheses, and the methodological approach to measure the features and the WTP.
Next, empirical results are shown. Finally, we discuss findings, highlight implications, and present some study limitations.
RELATED LITERATURE
Social Recommender Systems and Personalized News Aggregators
Personalization mechanisms such as recommender systems have been in existence since the introduction of the first system –
“Tapestry” – by Goldberg, Nichols, Oki, and Terry (1992). These technologies assist the user by supplying well-structured
information in searching, sorting, and filtering the massive amount of information available online. Initially used in e-
commerce (e.g. product recommendations by amazon.com), recommender systems can now also be used for digital products,
such as news or music. Different technologies have been developed; the traditional and most widely used ones are content-
based filtering, collaborative filtering, and hybrid filtering (Adomavicius and Tuzhilin, 2005).
With the rise of Web 2.0, social networks have spread and (inter)personal information has become available, for instance via
Facebook (Carmagnola, Vernero and Grillo, 2009). Based on information about a user or users’ social networking friends,
social recommender systems can recommend content (Ricci, Rokach, Shapira and Kantor, 2011). As various scholars note,
social recommender systems devise a new way to improve both the selection and the weighting of recommendations, thereby
increasing recommender systems’ accuracy and enable a new consumption of content (e.g., Arazy, Kumar and Shapira,
2010). These systems enable the automated selection, bundling, and combination of content from different sources, adapted
to an individual consumer’s preferences.
Initially, IT-enabled personalization mechanisms such as recommender system technologies were integrated into aggregation
systems. Based on recommender systems, they provided a first solution to the simple task of bundling content. Aggregation
systems add value by analyzing and adjusting information from different sources according to a specific objective (Zhu,
Siegel and Madnick, 2001). Based on the new technology of social recommender systems, PNAs are the new generation of
aggregation systems. Paliouras, Mouzakidis, Moustakas, and Skourlas (2008) explain a mechanism that aggregated content,
sorting them into categories and presenting an adaptively personalized interface. Nanas et al. (2010) illustrated, by means of a
self-developed PNA, how content-based filtering can be useful for selecting relevant information.
Business Model for News
Traditionally, news was bundled in a paper-based newspaper and has been sold at low prices to individuals or corporate
subscribers. Advertising has been sold in order to cover the costs and generate revenue. This established approach became
known as the “newspaper revenue model” (Teece, 2010). Owing to digitalization, publishers began to move from print
newspapers to an online form; however this has created monetization problems (Saeaeksjaervi et al., 2003). As noted, WTP
for online content is low, and it is unclear whether advertising can entirely compensate for the loss in direct revenue.
According to Chyi (2005), publishers have been experimenting with different business models for online news: the
subscription model, the advertising model, the transactional model, and the bundled model. For example, Wang, Ye, Zhang,
and Nguyen (2005) showed that several factors (e.g. added-value or service quality) influence the willingness to access
subscription based news. Additionally, “freemium” – as a new revenue model – has the potential to monetize news, since a
free version as well as a paid-based premium version (e.g. without advertisement) are provided (Wagner, Benlian and Hess,
2013). Moreover, research is starting addressing digital media innovations in order to find a solution for news delivery in the
future (e.g., Gershon, 2013).
RESEARCH MODEL AND HYPOTHESIS DEVELOPMENT
First, we address the development of a research model in order to answer RQ1. To do so, we want to examine user attitudes
to adopting a PNA and how this is influenced by different features. The research model of Dibbern, Heinzl, and Schaub
(2007), which analyzes determinants that affect mobile banking services acceptance, seems to be appropriate for this study.
The research model is based on the Technology Acceptance Model (TAM) and the Theory of Reasoned Action (TRA)
(Davis, 1989; Fishbein and Ajzen, 1975). The purpose of technology acceptance models is to provide a theoretical framework
to analyze a technology’s adoption. It argues that an individual’s behavioral intention to perform a certain action is the result
of his or her attitude towards this behavior, which leads to a specific use behavior. This procedure has been validated (e.g.,
Davis, 1989; Doerr, Benlian, Vetter and Hess, 2010).
Oechslein et al. Paying for News: Opportunities through PNAs
Proceedings of the Nineteenth Americas Conference on Information Systems, Chicago, Illinois, August 15-17, 2013. 3
In this case and according to literature, we examine whether the user attitude (AT) to use a PNA influences the behavioral
intention (BI) to use a PNA. It is therefore hypothesized that:
H1: User AT will have a positive influence on the BI to use a PNA.
In a second step, we identified features of a PNA that influence user attitude, and to integrate these into the research
framework. We will draw on hypotheses for each feature to analyze its influence on AT. To support the development of the
framework a qualitative study was conducted first, with the aim to confirm literature-based features and to discover
exploratory new features in the context of PNAs. The study was conducted in mid-2012, including 34 interviews with
technology experts, such as employees, bloggers, or journalists. The following features are the result: perceived usefulness
(PU), considering pricing (CP), usage comfort (UC), personalization (PE), ease of use (EU), system quality (SQ), platform
support (PS), source integration (SI), multimedia content (MC), trusted sources (TS), awareness (AW), social interaction
(SO), and social personalization (SP). Following Dibbern et al. (2007), we clustered these features according to three
different perspectives: consumer, technology, and network.
In Hypothesis 2, the consumer perspective will be consolidated. PU thereby refers to user beliefs that task performance is
efficient, and CP can be described by the underlying payment model and the relevance of occurring usage costs. UC refers to
simple and intuitive handling, while PE covers content personalization according to user interests. Hence, the following
hypotheses are formulated:
H2a: PU will have a positive influence on user AT to use a PNA.
H2b: CP will have a positive influence on user AT to use a PNA.
H2c: UC will have a positive influence on user AT to use a PNA.
H2d: PE will have a positive influence on user AT to use a PNA.
Hypothesis 3 describes features from a technology perspective. EU refers to the user beliefs that the use of the technology is
possible without much effort. SQ is explained by the reliability and presence of the presented information, while PS describes
the different types of devices that support PNA. SI can be explained by the ability to involve various and different sources.
MC represents the degree of combination of text, image, audio, or video file formats. Thus the following hypotheses are
formulated:
H3a: EU will have a positive influence on user AT to use a PNA.
H3b: SQ will have a positive influence on user AT to use a PNA.
H3c: PS will have a positive influence on user AT to use a PNA.
H3d: SI will have a positive influence on user AT to use a PNA.
H3e: MC will have a positive influence on user AT to use a PNA.
Finally, Hypothesis 4 describes a network perspective. TS refers to a source’s credibility and origin. AW relates to the extent
of brand popularity and recognition of the PNA. SO includes all forms of exchange, such as commenting or recommending,
between users. SP integrates the interests of users’ friends as a basis for better recommendations and selection. This can be
summarized in the following hypotheses:
H4a: TS will have a positive influence on user AT to use a PNA.
H4b: AW will have a positive influence on user AT to use a PNA.
H4c: SO will have a positive influence on user AT to use a PNA.
H4d: SP will have a positive influence on user AT to use a PNA.
Oechslein et al. Paying for News: Opportunities through PNAs
Proceedings of the Nineteenth Americas Conference on Information Systems, Chicago, Illinois, August 15-17, 2013. 4
Use behavior*
(UB)
Behavioral intention
(BI)
Attitude
(AT)
Perceived usefulness
(PU)
Considering pricing
(CP)
Usage comfort
(UC)
Personalization
(PE)
H2a
H2b
H2c
H2d
Ease of use
(EU)
System quality
(SQ)
Platform support
(PS)
Source integration
(SI)
Multimedia content
(MC)
H3a
H3b
H3c
H3e
H3d
Trusted sources
(TS)
Awareness
(AW)
Social interaction
(SO)
Social personalization
(SP)
H4a
H4c
H4d
H4b
H1
* The PNA as proposed does not exist, and it is not possible to explain the de facto use
behavior. Various scholars consider behavioral intention a reliable indicator for de
facto use behavior, whereas it is satisfactory to explain the behavioral intention
(Fishbein and Ajzen, 1975).
ConsumerTechnologyNetwork
Figure 1. Research framework: Features that determine user attitude
To answer RQ2 and to achieve a first impression of price sensitivity, we investigated user WTP for a PNA service. To
measure the WTP, different methods were applied in research, for example the conjoint analysis method and the Becker-
DeGroot-Marschack method (Miller, Hofstetter, Krohmer and Zhang, 2011). In this study, the price sensitivity meter (PSM)
of Van Westendorp (1976) seems appropriate, as it is highly suitable for the price estimation of innovative services. For
example, it has been validated for the pricing of music-as-a-service (MaaS) (Doerr et al., 2010).
RESEARCH METHODOLOGY
Measures
For part one of the study, to operationalize the research framework, validated constructs were used in the questionnaire.
Constructs were measured and rated on 7-point Likert scales, where 1 refers the lowest score and 7 the highest score. The
items for BI were adopted from Venkatesh, Morris, Davis, and Davis (2003), while items for AT were measured by a
semantic differential by Graf (2007). Items for the features were adapted and worded according to the scale by Sujan and
Bettman (1989).
To measure the WTP for part two of the study, four items of the PSM scale were used, to calculate the marginal cheapness
(MGP), the marginal expensiveness (MEP), the optimal price point (OPP), and the indifference price (IDP). In the study, we
pretended to use a monthly charge for the use of a PNA. Nevertheless, the PSM only measures the price consciousness, but
not the intention to buy or pay for the service (Van Westendorp, 1976).
Data Collection
The data for this study was collected using a quantitative standardized online survey. At the start of the survey, we showed a
short video explaining the PNA’s functionality, to ensure that all participants had the same knowledge base. A pretest was
conducted. The survey was developed with the software Unipark by Globalpark, and data was collected in January 2013.
Participants were invited with an invitation link sent via email to 5,030 students of a German university, via Facebook, and
Oechslein et al. Paying for News: Opportunities through PNAs
Proceedings of the Nineteenth Americas Conference on Information Systems, Chicago, Illinois, August 15-17, 2013. 5
via personal contacts. We collected 498 datasets, but we could only consider datasets from participants who had already been
using a PNA. Thus, our final sample comprised 116 valid datasets. We followed the usual approach of asking students in this
development stage and in similar use cases (i.e. recommender systems or MaaS) (e.g., Benlian, Titah and Hess, 2010; Chyi,
2005; Wagner et al., 2013). The participants’ age ranged between 18 and 54 years, whereas 85% were between 18 and 29
years old. 54% of the participants were male and 46% were female. The sample comprised 75% students, 20% employees
and 5% self-employed. Most of the participants (62%) rarely use a PNA, but 19% already use it weekly, 13% daily, and 6%
use a PNA several times a day.
RESULTS
For the first part of the study, structural equation modeling was used to test the hypotheses. Therefore, the software
SmartPLS 2.0 M3, using the partial-least-squares (PLS) algorithm, was used for all analysis (Ringle, Wende and Will, 2005).
The algorithm has the advantage of modeling latent constructs and predictive models, and is usable with small sample sizes
(Chin, 1998). Furthermore, PLS analysis is highly appropriate for our explorative study (Hair, Ringle and Sarstedt, 2011). In
this case, the software was used to calculate path coefficients and to determine the paths’ significance in the model (using
bootstrapping).
To analyze the quality of the model and provide a valid model, all values have to be above literature-based thresholds. All
items except SQ and PS have Cronbach’s α values above .06, which is acceptable in this early research stage (Henseler,
Ringle and Sinkovics, 2009). For SQ and PS, one indicator each was rejected. A new calculation of the model now showed
values above the threshold. Composite reliability shows values above .70 in all cases (Chin, 1998). Furthermore, the average
variance extracted (AVE) showed values above the threshold of .50 (Chin, 1998). Discriminant validity was analyzed by
comparing the latent construct correlation and the square root of the specific AVE. For each construct, the AVE’s value was
higher that the square root (Fornell and Larcker, 1981). Therefore, all constructs satisfied the reliability and validity criteria.
Perspective Construct Composite reliability
AVE Mean
BI .966 .904 5.155
AT .897 .634 5.389
PU .852 .659 6.276
CP .730 .503 6.412
UC .908 .767 6.014
Consumer
PE .889 .729 6.182
EU .912 .777 6.129
SQ .863 .760 6.458
PS .853 .746 6.097
SI .920 .793 6.294
Technology
MC .927 .810 4.849
TS .870 .696 5.677
AW .969 .912 3.383
SO .920 .793 4.157
Network
SP .953 .872 4.231
Table 1. Composite reliabilities, AVEs, and descriptive statistics
To analyze the structural model, we used Ball’s Q² as the indicator for predictive relevance, as well as Cohen’s effect sizes f²
and t-values to investigate the paths’ significance. Using the jackknifing procedure, Q² > 0 presents a predictive relevance of
the model, whereas Q² ≤ 0 suggests a lack of relevance. All constructs have a positive Q², indicating that we have predictive
relevance (Fornell and Larcker, 1981; Sarstedt and Wilczynski, 2009). In a second step, we analyzed f² to determine each
path’s effect size. A value of .02 indicated a small, a value of .15 a medium, and a value of .35 a large effect size (Cohen,
1988). All of our significant results showed at least a small effect size. Overall, our model and features can explain one-third
of the variance in user attitude to use a PNA (R² = .299). Furthermore, AT shows an R² of .400 to explain the variance in BI.
In total, 7 of our hypotheses can be supported. First, as expected, AT shows a positive influence on the BI to use a PNA,
supporting H1 (β = .632, p < .05). We found support for hypotheses H2a, H2b, H2c, and H2d, whereas PU, CP, UC, and PE
positively influence AT (β = .230, p < .05 / β = .188, p < .05 / β = .238, p < .05 / β = .192, p < .05). On the one hand, the only
significant relationship for the hypothesis 3 is SQ of a PNA (β = -.233, p < .05). But, a negative relationship leads one to
reject hypothesis H3b. On the other hand, H3a, H3c, H3d, and H3e also cannot be supported (β = -.079, p > .05 / β = -.063, p
> .05 / β = .081, p > .05 / β = .009, p > .05). Therefore, EU, PS, SI, and MC do not lead to a higher AT. AW and SP
positively influence AT (β = .208, p < .10 / β = .393, p < .01). Therefore, hypotheses H4b and H4d can be supported. Finally,
Oechslein et al. Paying for News: Opportunities through PNAs
Proceedings of the Nineteenth Americas Conference on Information Systems, Chicago, Illinois, August 15-17, 2013. 6
TS and SO are significant but show negative relationships (β = -.286, p < .01 / β = -.394, p < .01). Hence, hypotheses H4a
and H4c must be rejected.
Perspective Hypothesis t-value β-value Effect size Result
H1
+
9.181 .632 - Supported
H2a
+
2.040 .230 .067 Supported
H2b
+
2.085 .188 .039 Supported
H2c
+
2.346 .238 .046 Supported
Consumer
H2d
+
1.968 .192 .039 Supported
H3a
+
.783 -.079 -.016 Not supported
H3b
+
2.057 -.233 .039 Not supported
H3c
+
.936 -.063 .004 Not supported
H3d
+
.867 .081 .006 Not supported
Technology
H3e
+
.102 .009 .010 Not supported
H4a
+
3.084 -.286 .098 Not supported
H4b
+
1.809 .208 .034 Supported
H4c
+
2.958 -.394 .083 Not supported
Network
H4d
+
2.666 .393 .066 Supported
Table 2. Results of the structural equation model and hypothesis validation
To evaluate the WTP for part two of our research, we aggregated all data and calculated four price points, following Van
Westendorp (1976). Price indications to the participants are scaled in 0.50€ intervals steps, ranging from 0€ to 20€. Results
are shown in a diagram presenting four graphs, with price on the X-axis and the cumulative percentage of the participants on
the Y-axis. MGP is calculated by the intersection point of not cheap and too cheap, because users considering the service as
too cheap would exceed users determining the service as not cheap. Using two mathematical functions, MGP has been
calculated as 0.42€. Obtaining a pricing range for the PNA use, MEP can be calculated by the intersection of too expensive
and not expensive, and shows a price of 6.83€. Again, if the price would be higher, the amount of users considering the
service as too expensive would exceed the users considering it as not expensive. OPP is calculated by the intersection of too
cheap and too expensive, and results in 1.88€ for PNA use. IDP is calculated by the intersection of cheap and expensive,
resulting in a higher price than the OPP. Here, IDP is 2.83€, showing that 25% of the users consider it too cheap and 25%
consider it too expensive. However, 50% of the users consider it an acceptable price. In this case, there is a high difference
between the IDP and the OPP of 0.95€ (2.83€ - 1.88€), indicating that users consider a PNA service to cost more than they
are willing to pay for it.
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Cum. distribution
Price in €
not cheap not expensive too expensive too cheap
Range of acceptable prices
MGP
0.42
OPP
1.88
MEP
6.83
Figure 2. WTP: Price range and optimal price point
Oechslein et al. Paying for News: Opportunities through PNAs
Proceedings of the Nineteenth Americas Conference on Information Systems, Chicago, Illinois, August 15-17, 2013. 7
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Cum. distribution
Price in €
cheap expensive
IDP
2.83
Figure 3. WTP: Indifference price
CONCLUSION, IMPLICATIONS, AND LIMITATIONS
This study primarily sought to investigate the question whether a PNA service might contain the potential to establish a new
business model for news in the digital environment. First, we investigated usage features, influencing the adoption of the
service; second, we considered user WTP for the use of this service.
First, the results of our survey, showed that the PNA’s perceived usefulness, usage comfort, and awareness of a PNA raise
users’ intention to use the service. Considering pricing, the survey shows that the user will have a look at the service’s costs
before starting to use it. Last, the personalization of content and the possibility of social personalization are important
features for a PNA to have and positively influence intention to use. Contrary to our expectations, system quality, trusted
sources, and social interaction lead to lower intention to use a PNA. It shows that users do not want most current affairs all
the time and not according to the most trusted sources; what users want is the most suitable news according to his or her
preferences. Finally, platform support seems not to be an indicating feature, and neither does integrating different types of
sources, video, and music content. Ease of use also seems not to be a feature. Second, the exploration of the WTP and an
OPP of 1.88€ shows that users are willing to pay for the use of the service. The pricing of a monthly service is acceptable for
the user, even if news is available complimentary. Furthermore, the price range goes up to 7€ as a monthly fee, which might
be the basis for a sustainable business model. This study therefore contributes to current research by applying the research
framework and the PSM scale to a new research field and by extending the framework with different features.
Concerning the study results, PNAs provide the requirements to establish a new business model for news. A PNA’s
configuration should focus on the core functionality so as to simplify users’ consumption of news. Design, surface
appearance, and content appearance are important. Our results show that customization is a key feature of a PNA and might
lead to higher WTP. Personalization allows for customizing according to user preferences, as stated during usage, but social
personalization allows higher recommender accuracy and improves automatic aggregation of content. Since this is one of the
main improvements of PNAs, it promises a solid base for PNA development. PNAs can be applied in the environment of one
publishing house, to aggregate content from different media formats personalized for each user. It is not necessary to try to
sell one newspaper or magazine as a whole. The selection of interesting articles for a user can achieve higher revenues.
Selling articles selected by the underlying technology could easily be implemented by a PNA.
The study provides a first insight in a potential future business model. However, our study also has some limitations. First,
the sample consists mostly of students and is therefore not representative for PNAs users. Results for the WTP might be
biased using this sample, as students provide a lower purchase power. Future studies should make use of a representative
sample, in order to transfer the results of the study to a wider population. Second, it should be explored how the development
of mobile Internet and mobile technologies affect PNAs in future. The continuance and discontinuance of PNAs is also an
interesting research area. Hence, this study is only a snapshot in time and it should be replicated in the future. Third, the
research model was tested in only one special application context (i.e. PNAs). The model should be tested with other digital
Oechslein et al. Paying for News: Opportunities through PNAs
Proceedings of the Nineteenth Americas Conference on Information Systems, Chicago, Illinois, August 15-17, 2013. 8
services or products. Also, other (moderating) variables should be explored in future studies to help draw a more complete
picture of the investigated relationships and the affected user intention. However, we could explain 30% of the attitude.
Fourth, predictions about the WTP are affected by other influencing factors that are not addressed in this study. WTP should
be examined in future, for example, to investigate price sensitivity for different features of a PNA.
REFERENCES
1. Adomavicius, G. and Tuzhilin, A. (2005) Towards the next generation of recommender systems: A survey of the state-
of-the-art and possible extensions, IEEE Transactions on Knowledge and Data Engineering, 17, 6, 734-749.
2. Arazy, O., Kumar, N. and Shapira, B. (2010) A theory-driven design framework for social recommender systems,
Journal of the Association for Information Systems, 11, 9, 455-490.
3. Benlian, A., Titah, R. and Hess, T. (2010) Provider- vs. User-generated Recommendations on E-Commerce Websites –
Comparing Cognitive, Affective and Relational Effects, in Proceedings of the International Conference on Information
Systems, December 12-15, St. Louis, USA, ACM Press, 1-19.
4. Carmagnola, F., Vernero, F. and Grillo, P. (2009) Sonars: A social networks-based algorithm for social recommender
systems, in Proceedings of the seventeenth International Conference on User Modeling, Adaptation, and
Personalization, June 22-26, Trento, Italy, Springer-Verlag, 223-234.
5. Chin, W. W. (1998) The partial least squares approach for structural equation modeling, in George A. Marcoulides
(Eds.) Modern Methods for Business Research, Lawrence Erlbaum Associates, Hillsdale, 295-336.
6. Chyi, H. I. (2005) Willingness to Pay for Online News: An Empirical Study on the Viability of the Subscription Model,
Journal of Media Economics, 18, 2, 131-142.
7. Cohen, J. (1988) Statistical power analysis for the behavioral sciences, Lawrence Erlbaum Associates, Hillsdale.
8. Davis, F. D. (1989) Perceived usefulness, perceived ease of use, and user acceptance of information technology, MIS
Quarterly, 13, 3, 319-340.
9. Dibbern, J., Heinzl, A. and Schaub, N. (2007) Determinanten der Akzeptanz von mobilen Bankdiensten: Test eines Drei-
Perspektiven-Modells, in Tomas Bayon, Andreas Hermann and Frank Huber (Eds.) Vielfalt und Einheit in der
Marketingwissenschaft, Springer, Wiesbaden, 449-478.
10. Doerr, J., Benlian, A., Vetter, J. and Hess, T. (2010) Pricing of Content Services – An Empirical Investigation of Music
as a Service, in Proceedings of the sixteenth Americas Conference on Information Systems, August 12-15, Lima, Peru,
ACM Press, 13-24.
11. Dou, W. (2004) Will Internet users pay for online content?, Journal of Advertising Research, 44, 4, 349-359.
12. Fishbein, M. and Ajzen, I. (1975) Belief, attitude, intention, and behavior: an introduction to theory and research,
Addison-Wesley Pub. Co., Reading.
13. Fornell, C. and Larcker, D. F. (1981) Evaluating structural equation models with unobservable variables and
measurement error, Journal of Marketing Research, 18, 1, 39-50.
14. Gershon, R. A. (2013) Digital Media Innovation and the Apple iPad: Three Perspectives on the Future of Computer
Tablets and News Delivery, Journal of Media Business Studies, 10, 1, 41-61.
15. Goldberg, D., Nichols, D., Oki, B. M. and Terry, D. (1992) Using collaborative filtering to weave an information
tapestry, Communications of the ACM, 35, 12, 61-70.
16. Graf, D. (2007) Die Theorie des geplanten Verhaltens, in Dirk Krüger and Helmut Vogt (Eds.) Theorien in der
biologiedidaktischen Forschung, Springer, Berlin Heidelberg, 33-43.
17. Hair, J. F., Ringle, C. M. and Sarstedt, M. (2011) PLS-SEM: Indeed a silver bullet, The Journal of Marketing Theory and
Practice, 19, 2, 139-152.
18. Henseler, J., Ringle, C. M. and Sinkovics, R. R. (2009) The Use of Partial Least Squares Path Modeling in International
Marketing, Advances in International Marketing, 20, 277-319.
19. Miller, K. M., Hofstetter, R., Krohmer, H. and Zhang, Z. J. (2011) How Should Consumers' Willingness to Pay Be
Measured? An Empirical Comparison of State-of-the-Art Approaches, Journal of Marketing Research, 48, 1, 172-184.
20. Nanas, N., Vavalis, M. and Houstis, E. (2010) Personalised news and scientific literature aggregation, Information
Processing & Management, 46, 3, 268-283.
Oechslein et al. Paying for News: Opportunities through PNAs
Proceedings of the Nineteenth Americas Conference on Information Systems, Chicago, Illinois, August 15-17, 2013. 9
21. Paliouras, G., Mouzakidis, A., Moustakas, V. and Skourlas, C. (2008) PNS: A personalized news aggregator on the web,
Intelligent Interactive Systems in Knowledge-based Environments, 104, 175-197.
22. Ricci, F., Rokach, L., Shapira, B. and Kantor, P. B. (2011) Recommender Systems Handbook, Springer, New York.
23. Saeaeksjaervi, M., Wagner, M. and Santonen, T. (2003) Customization as a Business Model for Online Newspapers, in
Proceedings of the sixteenth Bled eCommerce Conference, June 9-11, Bled, Slovenia, ACM Press, 1-13.
24. Sarstedt, M. and Wilczynski, P. (2009) More for less? A comparison of single-item and multi-item measures, Die
Betriebswirtschaft, 69, 2, 211-227.
25. Sujan, M. and Bettman, J. R. (1989) The effects of brand positioning strategies on consumers' brand and category
perceptions: Some insights from schema research, Journal of Marketing Research, 26, 3, 454-467.
26. Teece, D. J. (2010) Business models, business strategy and innovation, Long Range Planning, 43, 2, 172-194.
27. Van Westendorp, P. H. (1976) NSS-price sensitivity meter (PSM)-A new approach to study consumer perception of
price, in Proceedings of the twenty ninth ESOMAR Congress, Amsterdam, Netherlands, ACM Press, 139-167.
28. Venkatesh, V., Morris, M. G., Davis, G. B. and Davis, F. D. (2003) User acceptance of information technology: Toward
a unified view, MIS Quarterly, 27, 3, 425-478.
29. Wagner, T., Benlian, A. and Hess, T. (2013) The Advertising Effect of Free-Do Free Basic Versions Promote Premium
Versions within the Freemium Business Model of Music Services?, in Proceedings of the forty six Hawaii International
Conference on System Sciences, Hawaii, USA, IEEE, 1-10.
30. Wang, C. L., Ye, L. R., Zhang, Y. and Nguyen, D.-D. (2005) Subscription to fee-based online services: what makes
consumer pay for online content, Journal of Electronic Commerce Research, 6, 4, 304-311.
31. Zhu, H., Siegel, M. and Madnick, S. (2001) Information Aggregation-a Value-Added E-Service, in Proceedings of the
fifth International Conference on Technology, Policy and Innovation, June 26-29, Delft, Netherlands, ACM Press, 1-11.
