Parents’ path to app purchase:
Examining the motivations and influences behind parental app
purchasing
Author: Danielle King Supervisor: Dr. Sanne Opree Student ID: 10583696 Master’s Thesis Master’s Program
Communication Science 30 January, 2015
Abstract
Recent media statistics show that parents have widely adopted the behavior of downloading mobile apps for the use of their children. However, scientific analysis of parental app purchasing is currently lacking and needed in order to better understand this behavior. This study’s three aims are to (1) gauge parents’ own experiences with downloading apps for their children; (2) examine how underlying behavioral factors (i.e., attitudes, perceived educational value, and perceived behavioral control) influence app purchasing; and (3) test how parents evaluate different marketing claims. A mixed-method design was implemented with 96 parents of children between the ages of 2-6. The first part of the study consisted of a survey constructed to gauge parents’ opinions and experiences with children’s apps. The second part implemented an experiment in order to analyze differences between groups following exposure to different marketing claims. As expected, results from the survey indicate that parents are regularly
downloading apps for their children and are very much involved in the process. It was found that parents take time to read app descriptions and consider them an important aspect of their
purchase decision. Additionally, it was found that general attitudes towards apps are the strongest predictor of parental app purchasing. Unexpectedly, it was found that perceived educational value of apps and perceived behavioral control do not influence parents’ app purchasing behavior. Results of the experiment indicate that parents express similar perceived educational value, attitudes, and purchase intent following both vague and direct marketing descriptions. Theoretical and practical implications are discussed.
Parents’ path to app purchase: Examining the motivations and influences behind parental app purchasing
One could argue that 1997 was a pivotal point in the field of media aimed specifically at young children. Two new programs were introduced during this year that greatly influenced the infant and toddler media market: the first was the launch of the television phenomenon
‘Teletubbies,’ the second was Walt Disney Company’s release of a video collection titled ‘Baby Einstein.’ This is not to say that children’s media did not exist prior to this time, but the launch of these programs signified an aggressive marketing push of media that specifically targeted parents of toddlers and pre-school aged children (Wartella, Richert & Robb, 2010; Brown, 2011). While ‘Teletubbies’ (and other similar shows that followed) was introduced mostly as entertainment, ‘Baby Einstein’ was specifically promoted as educational content. The ‘Baby Einstein’ video collection, that used music, shapes, colors and few words, was marketed as appropriate and valuable for the cognitive development of babies and toddlers and was a
phenomenal success among parents. Since the late 90’s there has been an explosive growth in the market of educational media and the use of such media by very young children (Deloache & Chiong, 2009; Vandewater et al., 2007).
The market has advanced to new media such as mobile applications, which has been exacerbated with the creation of the app store in 2008 and the introduction of the ‘kids’ section of the app store in 2013. App reception by the general public has been extremely successful with 75 billion downloads as of June 2014 (Perez, 2014). Surveys with parents suggest that apps for children are indeed a large part of this success as 68% of US parents have already purchased or intend to purchase an app for their 2-10 year old children (Dogtiev, 2014). Furthermore, Chiong and Shuler (2010) found that the majority of parents had downloaded an average of 20 apps on their personal devices for the use of their children with 7% of parents having more than 60 children’s apps on their phone or tablet. It remains largely unclear, however, what specifically is driving the immense success of children’s apps amongst parents. What are the intentions and motivations behind these purchases and what is influencing them?
The question of influence becomes crucial when considering the marketing of these apps and the public controversy regarding the ways in which baby media is promoted to parents (Vaala and Lapierre, 2013). Apps are presented in the app store with a description of content
which is written by the app’s producers in very much the same way as a baby DVD would have a description on the back of its cover. Most of these descriptions contain educational claims which have been defined as presenting the content in a way that implies it, “can assist children in learning important information, skills, values, and behavior while entertaining them and exciting their curiosity to learn about the world around them” (Fenstermacher et al., 2010, p.561).
Concerns were raised regarding the claims made on children’s educational DVD’s in 2006 when the Campaign for Commercial Free Childhood (CCFC) filed a complaint to the Federal Trade Commission (FTC) claiming that infant-directed media producers were misleading consumers with empty promises of educational outcomes that have neither been tested nor proven (Lewin, 2009). A similar sequence of events occurred in 2013 when the CCFC filed a complaint against baby app company Open Solutions (Bachman, 2013) for misleading parents to believe that using the company’s apps would lead to educational results.
In response to the complaints, some producers have changed their marketing claims from direct statements of educational outcomes to more general, and arguably vague, descriptions which were approved by the FTC. Policies such as these regarding media marketing are made based on various assumptions that have not yet been scientifically tested. In the case of
marketing apps for children these assumptions include the notion that parents regularly read app descriptions, and that they adjust their expectations based on whether the descriptions directly imply educational outcomes or only vaguely allude to them. Previous research done by Vaala and Lapierre (2013) found that when evaluating children’s DVD covers parents were unable to differentiate between direct and vague marketing claims. Research must be conducted in order to assess if the same effect occurs in the case of app descriptions.
Given the magnitude of children’s apps’ popularity and the ongoing debate regarding the policies concerning their marketing, it becomes clear that more research is needed to examine what exactly influences parents before they make the final decision to press the download button. This study poses three aims intended to minimize the gap in the literature regarding parents and their path to app purchases for their children. These aims are as follows: (1) gathering general information about parental app purchasing behavior (i.e., downloading free or priced apps); (2) investigating the underlying intentions and motivations behind this behavior and determining the most influential factors; and (3) examining the influence of direct and vague marketing claims on this behavior.
Theoretical Framework
General Information from Parents about App Purchasing
The first aim of this study is to gather general information about parents’ experiences, habits, and opinions related to the purchasing of apps for their children. As emphasized in the introduction, market research indicates that parents are enthusiastic purchasers of children’s apps. A study conducted by the Joan Ganz Cooney Center found that over 25% of all parents have downloaded apps for the use of their children (Shuler, Levine & Ree, 2012). Similarly, a survey done by PBS found that 7 in 10 parents indicated they has already purchased or plan to purchase apps for their kids (PBS Kids, 2013). The survey also found that parents are not buying apps at random but are rather gravitating specifically towards educational apps, with 77% of parents naming educational content as important in their purchasing decision. The increasing demand for such content has created a surge of educational apps for children. As of 2012, 80% of the apps in the education section of the app store directly targeted children, with the
toddler/preschool age being the most popular. Apps targeting young children have surpassed the popularity of adult apps indicating the incredibly strong market demand for this form of
children’s media (Lenhart, 2012). The above surveys and reports provide valuable information regarding the market popularity of children’s apps among parents. Yet, seeing as parents are the predominant purchasers of these wildly popular apps it is imperative to hear from them first-hand what considerations and decisions they make before their purchases.
This study aims to add to the literature by providing a more in depth understanding of parents’ thought process as they purchase these apps. While the existing literature provides sales figures and a basic notion of the type of content parents are looking for, this study hopes to uncover more specific details to complete the picture. These details include the ways in which parents are exposed to apps, whether they rely on reviews and descriptions, and what they consider the main reason for downloading apps. Furthermore, this study hopes to go beyond the purchase point and discover what happens after the app is downloaded (e.g., rules and
regulations parents enforce in their household specifically regarding apps) in order to better understand the process from beginning to end. This information can help illuminate other
findings in this study as well as serve as a basis for future research on this topic. Based on market research this study expects to find that children’s apps are in fact popular among parents and that
parents predominantly choose to download apps from the education category of the app store. It is also expected that parents will cite education as the main reason they download apps for their children.
Underlying Intentions and Motivations of App Purchasing
The second aim of this study is to examine parental behavioral intentions (using the integrated model of behavioral prediction described below) and determine the underlying
influential factors driving app purchasing among parents in general. Seeing as this behavior is so prevalent in society, it is of academic, as well as social, importance to test the cognitions that may be influencing parents to make their downloading decisions. It is crucial to understand which specific beliefs and/or opinions render a parent more or less likely to purchase an app for their child, or to become a frequent purchaser of children’s apps. Previous research has tried to establish what factors predict parents’ decisions to purchase certain media content for their children or regulate their children’s media use. Findings, however, are somewhat inconclusive.
On the one hand previous literature suggests that the main reason parents expose their children to media is that they believe it has educational value (Rideout, Vandewater & Wartella, 2003; Rideout & Hamel, 2006). This may explain why educational apps are the most popular apps among parents (FTC, 2012) and why 80% of parents in a recent survey indicated that educational content is something they look for when downloading an app for their child (Dogtiev, 2014).
On the other hand, previous research with parents also suggests that other factors such as general attitudes (i.e., positive or negative attitudes towards media not directly related to its perceived educational value) (Barr et al., 2010) and perceived behavioral control (Vaala & Hornik, 2013) are also associated with parents’ decisions regarding their children’s media use. This study aims at examining different influential cognitions and determining which are the most influential in predicting this behavior. To achieve this aim the study will implement the
integrated model of behavior prediction that includes all aforementioned factors and hence serves as an appropriate starting point.
The integrative model of behavioral prediction (Fishbein & Ajzen, 2010) is an extension to both the theory of reasoned action (Fishbein & Ajzen, 1975) and the theory of planned behavior (Ajzen, 1991) and is based on recognized well-established models for predicting a
variety of different behaviors (Sheppard, Hartwick & Warshaw, 1988; Vaala & Hornik, 2013). The model is the latest addition to these theories that evaluate the relationship between actions and intentions by taking into account the concepts of beliefs and attitudes as well as one’s perceived control over the intended behavior (Yzer, 2012). According to the model, other external factors such as demographics do not directly affect behavior but rather first influence a person’s beliefs, which then influences cognitions, and finally, intentions (Fishbein & Ajzen, 2010). The model explains that in order to predict a person’s behavior it is necessary to first evaluate whether that person intends to perform the behavior or not. The intention, therefore, starts from a person’s external factors that influence personal beliefs. These beliefs then
influence cognitions such as attitudes and perceived behavioral control that finally influence the intention to perform the behavior and the actual execution of the behavior (see Figure 1). In this way, perceived behavioral control is influenced by the belief in the actual ability to perform the behavior while attitudes are influenced by beliefs regarding the outcome of the behavior. However, this study will differentiate between attitudes towards apps in general and beliefs regarding apps’ educational value as the aforementioned previous research has indicated that these are indeed independent factors influencing parental media decisions. In other words, perceived educational value (i.e., beliefs in a certain outcome of app use) will be considered as an independent factor directly influencing intentions to perform the behavior (i.e., app
purchasing).
While the original model contends that perceived societal norms (pressure to adhere to social expectations) also influence intention, this study aims to focus solely on the individual behavioral process of app purchasing. Therefore, it will not focus on external factors such as societal influences as similar studies (Vaala & Hornik, 2013) have found perceived societal norms to be only marginally significant in predicting parental behavior in regards to the media use of their children. Therefore, the model used in this study will test the influence of general attitudes, perceived educational value and perceived behavioral control on app purchasing behavior. The next part of this section will elaborate on these factors and introduce their
corresponding hypotheses. The hypotheses in this section will be examined and tested by means of survey questions.
Figure 1
The Integrated Model of Behavioral Prediction (adapted from Fishbein & Ajzen, 2010).
Perceived Educational Value
Previous research has indicated the perceived educational value of media influences parents in their decisions regarding their children’s media use. A study done by Barr et al. (2010) found that 29% of parents believed television to be educationally beneficial to their children which led them to allow their kids to watch TV. Moreover, a study done by the Joan Ganz Cooney Center found that parents of children that frequently used mobile apps believed their children learn “a lot” from the apps (Rideout, 2014). As the integrated model of behavioral prediction suggests, beliefs of positive outcomes influence the intention to perform behaviors. In the case of beliefs regarding the perceived educational value of apps, parents may consider a measurable improvement in a certain skill such as speaking or counting a favorable outcome and be positively influenced to purchase apps. Thus hypothesis 1 is proposed:
H1: Parents that express general beliefs that apps are educational will demonstrate higher app purchasing behavior than parents that do not express general beliefs that apps are educational.
Attitudes
Attitudes in the integrated model of behavioral prediction refer to a person’s evaluation of the intended behavior and whether it is deemed favorable or unfavorable (Yzer, 2012). A study conducted by Vaala and Hornik (2013) that used the model to investigate the relationship between mothers and their children’s television viewing behavior found that amongst all factors within the model, the mothers’ attitudes were the strongest influence on the children’s TV habits.
Individual Variables e.g., demographics Beliefs regarding outcome Beliefs regarding control Behavior Intention Attitudes Perceived Control
The more positive mothers were about television the higher the viewing rates of their children. Such attitudes are not necessarily linked to any perceived educational value but to a positive or negative outlook on media in general. It is reasonable to assume that parents may consider their child being entertained or having fun a favorable outcome regardless of educational outcomes. Hence hypothesis 2 is presented:
H2: Parents that express positive attitudes towards apps in general will demonstrate higher app purchasing behavior than parents that do not express positive attitudes towards apps in general.
Perceived Behavioral Control
Perceived behavioral control was added to the theory of reasoned action that was then termed the theory of planned behavior (Ajzen, 1991). Perceived behavioral control thus refers to a person’s perception of whether or not the intended behavior can be successfully and fully carried out. Studies have shown that when people believe they have the ability to perform a behavior they are more likely to engage in that behavior (Arvola et al., 2008; Pavlou & Fygenson, 2006).
However, in terms of parental media behaviors, Vaala and Hornik (2013) suggested that control applies not only to the physical ability to carry out the behavior (accessing media) but also to the perceived control over the use of said media. Control over use could, for example, be expressed by parental mediation of how much time the child spends with the medium or rules implemented by parents to control how the child uses the medium in terms of accessing certain content.
While children are still very young (i.e., under age 6) parents act as media ‘gate-keepers’ who have control over which media is accessed and used (Barr et al., 2010), especially in
handheld devices that belong to the parents themselves. Parents hold most of the power over the experience their children have with these devices that are often handed over from parent to child for a controlled period of time (Chiong & Shuler, 2010). It is likely that parents’ beliefs
regarding whether or not they can control how their children will use the apps (in terms of the amount of time they spend with the apps and the content they are exposed to within the apps) will influence app purchasing among parents. Parents that are concerned about their children
becoming excessive app users or being exposed to inappropriate content may need a sense of control over their children’s app use before purchasing apps for them.
Furthermore, considering the parents’ role in selecting apps for their children to use, perceived control could also be applied to the level of control parents believe they have over the ability to select appropriate apps for their children. This control could be influenced by parents’ belief that they have the capabilities to judge apps and select those the child would both benefit from and enjoy. The following hypotheses address both perceived control over app use (H3) and perceived control over app selection (H4).
H3: Positive beliefs about parents’ ability to control how their child uses apps should positively influence app purchasing behavior.
H4: Positive beliefs about parents’ control over app selection should positively influence app purchasing behavior.
The Influence of Educational Marketing Claims on App Purchasing
The third and final aim of this study is to determine how educational claims made in app descriptions influence parents in their app downloads. Research has shown that the vast majority of educational media directed at young children does in fact market itself as developmentally constructive and educational (Vaala et al., 2010; Courage & Howe, 2010). The validity of such marketing has sparked much public controversy, which has led some media producers to alter their marketing from explicit statements to implicit ones. Fenstermacher et al. (2010) explain that explicit claims use “behaviorally specific verbs (e.g. teach, instruct) and specific behaviors or educational domains such as ‘teaches number recognition and order for numbers 6 to 10” while implicit claims “imply learning goals using non-specific language including verbs such as ‘explore’ and ‘introduce’ paired with a specific behavior or educational domain (e.g. ‘inspiring early language development—from simple gestures to first spoken words’)” (p.562). Implicit descriptions are, in a way, considered the solution to this debate as the FTC argued in 2007 that, “claims that a product ‘exposes’ or ‘introduces’ children to particular content—are unlikely, by themselves to convey an educational or developmental benefit claim…” (Engle as cited in Vaala & Lapierre, 2013, p.19). It is not entirely clear, however, whether parents are able to detect such differences in claims when evaluating media for their children.
In order to investigate how parents perceive differences in the wording of marketing claims in the case of apps, this study will elaborate on the extensive work done by Vaala and Lapierre (2013) whom compared parental exposure to explicit and implicit children’s DVD descriptions. In their study, the researchers tested and compared how the different claims affected attitudes towards the DVD, the perceived educational value of the DVD, and parents’ purchase intent of the DVD. All comparisons yielded insignificant differences meaning parents were not able to differentiate between direct and vague claims and were influenced in the same way by both types of marketing. Based on these results, similar outcomes are expected following exposure to direct and vague marketing claims in app descriptions. These assumptions will be tested using an experiment in which parents will be exposed to either direct or vague marketing claims.
H5: There will be no difference of the perceived educational value of the app between the parents exposed to a vague app description and the parents exposed to a direct description.
H6: Parents will express the same brand attitudes towards both an app with a vague description (“this will teach your child”) and an app with a direct description (“this will inspire your child”).
H7: Parents will express similar purchase intent for both an app with a vague description (“this will teach your child”) and an app with a direct description (“this will inspire your child”).
Methods
Sample and Procedure
The study implemented a mixed-method design combining both a cross-sectional survey and a between-subject experiment. Participants received an email that provided a general
description of the study and information regarding their right to withdraw from the study at any time. Participants were also promised that their anonymity would be safeguarded. The email provided a link to an online questionnaire that took 10-15 minutes to complete. Parents were told in the email that they should participate in the study only if they have children between the ages of 2-6, and if they have access to the appropriate technology (i.e., own at least one smartphone and/or tablet with internet access). This was done to ensure that respondents could in fact access apps on a regular basis and were familiar with the app store and the process of downloading apps. Ultimately, data was collected from 96 parents who were recruited using non-probability
snowball sampling via social media as well as direct invitations at the British School of
Amsterdam. Parents were asked to recommend other possible participants whose children meet the age criterion. The age of the parents (i.e., the respondents) ranged from 27 to 51 (M = 36, SD = 4.32) with children representing all ages between 2 and 6 (M = 3.78 SD = 1.38). Respondents were mostly highly educated with 76% holding at least a Bachelor’s degree. The majority of respondents originated from the Netherlands (40%), Israel (22%), and The United States (19%). Other countries included: Spain, the United Kingdom, Italy and Switzerland.
The first part of the online questionnaire consisted of questions pertaining to participants’ general opinions and attitudes towards apps for children as well as their experience downloading such apps. This part was designed to collect information relating to the first two aims of the study. Parents were asked whether they had ever downloaded an app specifically for their child. Those that answered affirmatively continued to the rest of the questionnaire while those who answered negatively were directed to the end of the survey and excluded from the study. A total of 8 participants were excluded for this reason. Parents were told that if they have more than one child within the 2-6 age range, to think of one child and answer questions only regarding that child.
The second part contained an experimental manipulation in which participants were randomly assigned to one of two groups. This manipulation was designed to test the influences of different marketing claims to reach the third aim of the study. Each group was exposed to an app description page designed specifically for this study (see Figure 2). Participants were asked for their thoughts and opinions of this new app, how educational they perceived the app to be, and whether they would purchase it for their own child. The page was designed to resemble other app description pages for this age group in the app store. Upon reviewing multiple popular existing apps, it was decided that the app would feature an animal as a main character (in this case, a dog named Coco) and would promote the learning of numbers and letters. Other than the text the two images were identical. The text of the description served as the manipulation (direct and vague marketing claims), which was written based on Vaala and Lapierre’s (2013) study of DVD covers. The claims were modified based on current independent research of descriptions of popular apps in the ‘Kids 5 & under’ section of the app store (e.g. Toca Town, Peekaboo Kids and Playtime with Dora the Explorer.).
Direct Marketing Description Vague Marketing Description
Figure 2
Experimental Stimuli.
Measures
The measures used in this study will be presented according to their corresponding aim. Seeing as this specific topic of research has yet to be studied, existing scales used in previous research needed to be adapted to adequately measure the variables. Most scales implemented in this study were altered because they originally related to television use or to DVD’s whereas this study is concerned with apps. The factorial structure of the adjusted scales was assessed as well as their reliability by conducting confirmatory factor analyses (CFA) and computing their
Cronbach’s alpha. A proper scale would yield an eigenvalue higher than 1 and a Cronbach’s alpha of at least .70.
General Information from Parents about App Purchasing
The first series of questions in the survey were designed to gauge a better understanding of app purchasing behavior in general. The questions were presented in both quantitative and qualitative form and included questions regarding, for example, how parents are first exposed to the apps they purchase (parents were asked to rank the option from most important to least important), what they state as the main reason for purchasing apps for their children (ranking scale), whether they read descriptions and reviews (1 = Always, 5 = Never), how important they consider descriptions and reviews to be in their purchasing decisions (1 = Extremely Important, 5 = Extremely Unimportant) and what kind of household rules apply in terms of their children’s app use (open question).
App purchasing behavior. To measure app purchasing behavior parents were asked how
many apps they have downloaded in the past 6 months specifically for the use of their child. Response categories were as follows: 1 (0-5 apps), 2 (5-10 apps), 3 (10-20 apps) and 4 (over 20 apps). The variable was recoded into two categorical variables to indicate purchases of 5 or more apps or and of 10 or more apps. This was done to differentiate between moderate app purchasers and frequent app purchasers. This variable was used as the dependent variable for the analysis of the second aim of the study.
Underlying Intentions and Motivations of App Purchasing.
Perceived Educational value. To measure whether parents believe apps have educational
value in general, a 5-point scale adapted from Vaala and Lapierre (2013) was used. The items were changed so that they refer to app use and not DVD use as they originally stated. The scale consisted of three statements: ‘I believe apps can be educational,’ ‘I believe apps can teach my child new skills,’ and ‘I believe apps can teach my child new skills.’ Participants were asked how strongly they agreed with the statements with 1 representing strongly disagree (or strong beliefs that apps cannot educational) and 5 representing strongly agree (or strong beliefs that apps can be educational). A principal component factor analysis (PCA) was conducted which showed that the three statements form a single scale with only one component having an eigenvalue above 1
(eigenvalue 2.67). The component explains 88.99% of the variance in beliefs of whether apps are educational. The statements were computed into a scale that proved to be very reliable as its Cronbach’s alpha was .93 (M = 4, SD = .78).
Attitudes. To measure parents’ general attitudes towards purchasing apps for their
children a 5-point scale adapted from Vaala and Hornik (2013) was used after changing the words ‘letting children watch television’ to ‘letting children use apps.’ Participants were asked to state how strongly they agree or disagree with the following statements: “I believe it is wise to let children use apps in general,” “I believe letting children use apps may be beneficial to the child,” and “I believe letting children use apps in general may be harmful to the child.”
Items were recoded so that a score of 1 indicated a negative attitude towards apps while a score of 5 indicated a positive attitude towards apps. A principal component factor analysis (PCA) was conducted which showed that the three statements form a single one-dimensional scale with only one component having an eigenvalue above 1 (eigenvalue 2.07). The component explains 69.15% of the variance in attitudes towards apps. The statements were computed into an attitudes scale, which proved to be reliable as its Cronbach’s alpha was .80 (M = 3.75, SD = .78).
Perceived Behavioral Control of app use. To measure the control parents believe they
have over their child’s use of apps respondents were asked how strongly they agree or disagree with the following statements adapted from Vaala and Hornik (2013): ‘I believe I can control how much time my child spends with apps’ and ‘I believe I can control what content my child is exposed to within apps’ (r = .64). Responses were recoded so a score of 1 represented low perceived control over app use and a score of 5 represented high perceived control over app use. The items were computed to create ‘control over app use’ scale (M = 3.97, SD = .84, α = .78).
Perceived Behavioral Control of app selection. To measure the control parents believe
they have over their ability to select appropriate apps for their children, respondents were asked how strongly they agree or disagree with the following statements: ‘I believe I can select appropriate apps for my child’ and ‘I believe I know what kind of apps my child will enjoy using’ (r = .79). Responses were recoded so a score of 1 represented low perceived control over app selection and a score of 5 represented high perceived control over app selection. The items were computed to create ‘control over app selection’ scale (M = 4.06, SD = .84, α = .73).
The Influence of Educational Marketing Claims on App Purchasing
Marketing Claims. Participants were exposed to one of two app descriptions; a direct
description containing explicit educational claims regarding educational outcomes (N = 47) and a vague description containing implicit educational claims (N = 49). The direct marketing
description included the verb “teaches” implying a direct educational outcome of using the app. The direct claim also explicitly stated the number of words (“100 words”) and numbers
(“numbers 1-10”) the app would teach its users. The vague marketing claim included the word “inspires” which suggests a vague educational outcome. It did not mention what specifically will be taught but rather used phrases such as “expose kids to words” to convey the outcome (see highlighted text in Figure 2).
A manipulation check was conducted to test the effect of the experimental manipulation (i.e., the different descriptions) on parents’ perceived educational value of the app. The
manipulation check consisted of two questions regarding the educational outcomes in the app descriptions. Participants were asked how many words and how many numbers they believe their child will learn from the app to determine a difference between experimental groups. It was expected that if parents read the descriptions carefully those exposed to the direct description would remember the exact number of words and numbers and would be more likely to name the correct answer than those exposed to the vague description.
An independent-samples t-test was conducted to compare how many numbers parents believed their child would learn between the group exposed to the direct marketing app description and the group exposed to the vague marketing app description. A significant difference was found between the vague group (N =39, M = 5.56, SD = 4.41) and the direct group (N = 44, M = 7.73, SD = 4.06); t (81) = -2.32, p = .023, 95% CI [-4.01, -.31]).A second independent-samples t-test was conducted to compare the number of words parents believed their child would learn between the group exposed to the direct marketing app description and the group exposed to the vague marketing app description. A significant difference was found between the vague group (N = 49, M = 16.45, SD = 21.54) and the direct group (N = 47, M = 76.79, SD = 38.72); t (94) = -9.49, p<.001, 95% CI [-72.97, -47.70]). There was a significant difference in both how many words and how many numbers parents believed their child would
learn from the app (i.e., parents that were exposed to the direct description were more likely to name the correct answers). Therefore, the manipulation was successful.
Before testing the difference between means, both items (i.e., number of words and numbers parents believed their child would learn) were tested for outliers using a boxplot.
Outliers were not found for the items ‘words expected to be learned,’ but were found for the item ‘numbers expected to be learned.’ The 13 outliers consisted of values greater than 15 and were coded as missing values.
Interestingly, additional t-tests revealed that parents who were exposed to the vague app description indicated a lower perceived educational value of the app seeing as they believed the app would teach their child less words (N = 49, M = 16.45, SD = 21.54) and less numbers (N = 39, M = 5.56, SD = 4.41) than the parents exposed to the direct description (N = 47, M = 76.79,
SD = 38.72; N = 44, M = 7.73, SD = 4.06) (t (94) = -9.49, p<.001, 95% CI [-72.97, -47.70]; t (81)
= -2.32, p = .023, 95% CI [-4.01, -.31]).
Perceived Educational value. A single item was used to measure perceived educational
value as done by Vaala and Lapierre (2014). Parents were asked to rate the app based on its description page in terms of how educational they believed it to be. Responses were recorded using a 7-point scale (1 = not at all educational, 7 = very educational; M = 5.28, SD = 1.27).
Attitudes. An 11-point scale was adapted from Vaala and Hornik (2013) to refer to apps
and not television viewing. The scale was used to measure parents’ attitudes towards the app presented in the study. Respondents were asked to rate if the app presented to them is, for example, good/bad, pleasant/unpleasant, beneficial/harmful, and such. A scale was constructed by counting the number of positive words respondents selected, hence a score of 0 represents a very negative attitude towards app use and a score of 10 represents a very positive attitude towards apps (α = .88, M = 7.12, SD = 2.02).
Purchase intent. Parents were asked using a single item how likely they would be to
download the app for their child based on the description presented to them, as done by Vaala and Lapierre (2014). Responses were recorded using a 5-point scale (1 = very unlikely, 5 = very
Results
Analytical Plan
Each aim in this study was tested individually according to the variables involved and will be presented accordingly. Aim 1 concerned general information regarding app purchasing behavior and will therefore be presented with descriptive statistics of items such as the number of apps purchased in the past 6 months, the way in which parents are first exposed to apps, and such. The hypotheses in aim 2 involved continuous independent variables (i.e., attitudes, perceived educational value, and perceived behavioral control) and their influence on a dichotomous dependent variable (i.e., app purchases) and were therefore tested using logistic regressions. The correlation matrix (see Table 1) did not indicate any signs of multicollinearity (correlations larger than .50), so the independent variables could be combined into a regression model. The hypotheses in aim 3 involved exposing participants to an experimental stimulus (i.e., direct or vague app descriptions) and testing the differences among groups in terms of attitudes, perceived educational value, and purchase intent. Independent-sample t-tests were conducted to test these hypotheses by comparing the means of the dependent variables among the groups. General Information from Parents about App Purchasing
Participants were asked multiple questions about their opinions, experiences, and
behaviors regarding app purchases for their children. Parents were asked how they are generally first exposed to apps they download for their children; 50% of respondents indicated they find the apps themselves by browsing the app store, 30% indicated they found apps based on recommendations from friends, 10% find apps on online sources such as forums, and the
remaining 10% indicated they use professional app reviews, recommendations from teachers, or opted for the ‘other’ option. The option allowed participants to fill in their own answer; all parents that opted to fill in their own answer specified the child finding the app and asking for it directly.
After establishing how parents find the apps they download, participants were asked to rank reasons for downloading children’s apps according to their importance in order to gauge the main motivation behind the behavior. Answer items included: ‘they help educate my child’, ‘they entertain my child’, ‘they make my child happy’, and ‘they keep my child busy’. Education
was the option chosen by most parents as the most important reason they download apps for their children while keeping the child busy was chosen by the largest number of participants as the least important (see Table 2).
Additionally, parents ranked the category within the children’s section of the app store that they downloaded apps from most frequently (with first ranking being the category they download from most and the fourth raking being the category they download from least). Most parents indicated that they download apps mostly from the education category while the entertainment category is the category parents download the least from (see Table 3).
Next, parents were asked their opinions regarding app descriptions and app reviews presented in the app store. They were asked how often they read descriptions (1 = Always, 5 =
Never; M = 1.96, SD = 1.05) and reviews (1 = Always, 5 = Never; M = 2.45, SD = 1.04), and how
important they believe descriptions (1 = Extremely Important, 5 = Extremely Unimportant; M = 1.97, SD = .86), and reviews (1 = Extremely Important, 5 = Extremely Unimportant; M = 2.24,
SD = .75), are in their decision to buy an app for their child. While descriptions and reviews were
both found to be crucial, overall descriptions were found to be more important to parents in their decision to purchase an app for their child.
After covering the main aspects of the purchasing process, questions turned to the use of apps in the household. Participants were asked questions regarding the rules they enforce
regarding their children’s app use. Many respondents (76%) indicated they limit their child’s app use to a certain amount of time per day; of these parents, 46% indicated there is difference between the amounts of time their child is allowed to use apps on a weekday as opposed to the weekend. Answer options ranged from 1 (0-5 minutes per day) to 4 (more than 60 minutes per day). It seems that children are allowed to use apps for more time during the weekend. The option most chosen by parents for weekday use was 10-15 minutes, while the option most chosen for weekend use was 30-60 minutes.
Lastly, parents were asked if there are any other rules in their household regarding app use in the form of an open question. Some parents indicated, for example, that apps were allowed only at certain times of the day such as after 5:00 p.m. or “in the evening,” or if the weather did not allow for outside play.
Table 1
Zero-Order Correlations of Main Variables
Note. *p < .05. ** p < .01 (two-tailed). Educational Scale Attitudes Scale Control over use scale Control over content scale More than 5 apps More than 10 apps App groups Brand attitudes scale of app Purchase intent of app Educational value of app Educational Scale __ Attitudes Scale .37** __ Control over use scale -.14 .27** __ Control over content scale -.20* .41** .58** __ More than 5 apps -.12 .24* -.09 .08 __ More than 10 apps -.007 .21* -.08 .11 .46** __ App groups .12 .09 .03 -.005 -.18 -.07 __ Brand attitudes scale of app .48* -.36 -.47* -.20 .10 .17 .33 __ Purchase intent of app -.17 .25* .12* .34** .20 .04 -.14 -.77** __ Educational value of app -.34** .32** .19 .29** .24* .20 -.08 -.78** .55** __
Table 2
Parents’ Rankings of Reasons to Download Apps for Children
1 (Most Important) 2 3 4 (Least Important) Educate child 40% 26% 24% 10% Entertain child 22% 35% 30% 13%
Make child happy 21% 36% 31% 12%
Keep child busy 10% 6% 12% 72%
Table 3
Parents’ Rankings of Favorite Category within the App Store
1 (Most downloaded) 2 3 4 (Least downloaded) Education 53% 28% 14% 5% Games 35% 21% 35% 10% Books 2% 20% 42% 8% Entertainment 10% 21% 40% 28%
Others indicated that the use of apps was a reward for good behavior during the day (e.g. completion of homework) so it varied from day to day; while others explained that app use was only allowed under the supervision of an adult. Some parents also mentioned that the time
allowed with apps depended on the content of the app itself. As one parent explained: “How long he can use it depends on what app he is using, for instance, if it is a book/reading app then he can use it for more time than a game app.” Overall, parents indicated a certain level of control over app use within their household as well as control over the content of the apps used by their children; 65% of parents indicated that they always or almost always test apps themselves before allowing their children to engage with them (1 = Always, 5 = Never; M = 2.16, SD = 1.23).
Underlying Intentions and Motivations of App Purchasing
Two direct logistic regressions were performed to determine the influence of several factors on whether parents are likely to buy apps for their children. App purchasing behavior was recoded into two categorical variables to indicate purchasing 5 or more apps (Model 1) or
purchasing 10 or more apps (Model 2). Both models contained 4 independent variables (attitudes towards apps, perceived educational value of apps, perceived behavioral control over app use and perceived behavioral control over app selection) to test hypotheses H1-H4 (see Table 4). Model 1 was statistically significant, X2 (7, N = 96) = 9.81, p = .044, indicating that the model was able to distinguish which parents bought 5 or more apps and which did not (i.e., moderate purchasers). The model as a whole explained between 9% (Cox and Snell R square) and 13% (Nagelkerke R squared) of the variance in the purchase of 5 or more apps, and correctly classified 64.6% of the cases. As shown in Table 4, one of the independent variables made a unique statistically significant contribution to the model (attitudes towards apps; p = .044) while another independent variable proved to be marginally significant (control over app use; p = .064). Attitudes, which is the strongest predictor of purchasing 5 or more apps, recorded an odds ratio of 2.01 indicating that when attitudes are increased by one unit parents are twice as likely to purchase 5 or more apps, controlling for all other factors in the model. Control over app use recorded an odds ratio of .53; because the odds ration value is smaller than 1, parents with higher values of control over app use are less likely to buy 5 or more apps.
Model 2, containing all predictors was marginally significant, X2 (4, N = 96) = 8.59, p = 072. The model as a whole explained between 8% (Cox and Snell R square) and 12%
(Nagelkerke R squared) of the variance in the purchase of 10 or more apps, and correctly classified 75% of the cases. As shown in Table 4, attitudes towards apps (p = .049) made a unique statistically significant contribution to the model while control over app use made a marginally significant contribution (p = .069). Attitudes, which is the strongest predictor of purchasing 10 or more apps, recorded an odds ratio of 2.14 indicating that when attitudes are increased by one unit parents are twice as likely to purchase 10 or more apps, controlling for all other factors in the model. Control over app use recorded an odds ratio of .54; because the odds ration value is smaller than 1, parents with higher values of control over app use are less likely to buy more 10 or more apps.
Table 4
Logistic Regression Predicting Likelihood of Purchasing 5 or more Apps (Model 1)
B S.E. Wald df p Odds Ratio 95.0% C.I. of Odds Ratio Lower Upper Educational Value -.14 .30 .22 1 .63 .86 .47 1.57 Attitudes .69 .34 4.07 1 .04 2.01 1.02 3.96 Control over app use -.62 .33 3.42 1 .06 .53 .27 1.03 Control over app selection .27 .35 .61 1 .43 1.32 .66 2.63 Constant -.68 1.74 .15 .69 .50
Logistic Regression Predicting Likelihood of Purchasing 10 or more Apps (Model 2)
Attitudes were proven to positively predict app purchasing hence H2 is supported. Perceived behavioral control over app use was proven to negatively predict app purchasing thus negating H3, which is therefore rejected. Perceived educational value (H1) and perceived behavioral control over app selection (H4) were not proven to be significant predictors of app purchasing as the hypotheses predicted. Therefore, these are rejected.
Educational Value .22 .32 .48 1 .48 1.25 .66 2.35 Attitudes .73 .38 3.88 1 .04 2.14 1 4.57 Control over app use -.60 .33 3.29 1 .06 .54 .28 1.04 Control over app selection .42 .37 1.28 1 .25 1.52 .73 3.18 Constant -3.60 1.95 3.42 .06 .02
The Influence of Educational Marketing Claims on App Purchasing
The third and final aim of this study was to examine how parents evaluate apps based on different app descriptions, and how these descriptions influence perceived educational value, attitudes, and purchase intent. Parents in the sample were randomly assigned to one of two groups and were exposed to an app description using either direct educational marketing claims or vague educational marketing claims. The first independent-samples t-test was conducted to compare the perceived educational value scores between the group exposed to the direct marketing app description and the group exposed to the vague marketing app description. No significant difference was found for the direct group (M = 5.17, SD = 1.38) and the vague group (M = 5.39, SD = 1.15); t (94) = .837, p = .405, 95% CI [-.29, .73]). There was no difference in how both groups rated the educational value of the app, thus H5 is supported.
A second independent-samples t-test was conducted to compare brand attitudes between the group exposed to the direct marketing app description and the group exposed to the vague marketing app description. No significant difference was found for the direct group (M = 6.92,
SD = 2.26) and the vague group (M = 7.31, SD = 1.75); t (94) = .94, p = .347; 95% CI [-.42,
1.21]). There was no difference in brand attitudes towards the app among groups hence H6 is supported.
A third independent-samples t-test was conducted to compare purchase intent between the group exposed to the direct marketing app description and the group exposed to the vague marketing app description. No significant difference was found for the direct group (M = 3.34,
SD = 1.12) and the vague group (M = 3.65, SD = .96); t (94) = 1.45, p = .148; 95% CI [-.11,
.73]). There was no difference in purchase intent of the app among groups, thus H7 is supported. Discussion
This study’s aim was threefold as it was designed to gain an understanding of the process parents undergo in purchasing apps for their children, the influential underlying behavioral factors driving this process, and the influence of promotional educational claims. This study was the first to profoundly research this specific media behavior among parents of young children. Overall, findings suggest that parents are mostly influenced by their attitudes towards apps in general when it comes to purchasing apps for their children. Additionally, it was found that although parents predict different educational outcomes (i.e., predicted words and numbers their
child will learn) from apps with vague marketing and apps with direct marketing claims, there is still no difference in their overall educational evaluation of such apps, their attitudes towards the apps and their purchase intent of the apps.
General Information from Parents about App Purchasing
The first aim of this study was to lay the groundwork for future research regarding parents’ experiences, opinions, and habits of app purchasing for their children. This general information regarding this behavior is mostly lacking and is crucial for research on this topic. In line with current media statistics (Dogtiev, 2014), findings suggest that children’s apps are popular among parents of young children with the majority of parents in the study having downloaded more than 5 apps for their child in the last 6 months. It seems that parents mostly buy apps for their children because they believe them to be educational. This is evident from reports of parents indicating that the main reason they buy apps is their belief in apps’ ability to help educate their child and that they purchase apps predominantly from the education category of the app store. These findings corroborate previous market research that has indicated that educational content is important to parents and that the education category is the most popular among parents of young children (Shuler et al., 2012; PBS Kids, 2013). Furthermore, this study presents new findings that suggest that app purchasing is a process parents are very much involved in and one they do not take lightly. Participants in this study indicated they take the time to search for apps themselves, read descriptions as well as reviews, test the apps before allowing their child to engage with them, and set rules and regulations regarding use. It is noteworthy that most parents indicated that they consider app descriptions to be important in their decision to purchase a certain app. This finding reinforces the need for additional research on the ways in which app developers are describing their apps and how these descriptions influence parents. Overall, these findings lay groundwork for this study’s additional findings as well as for future research into parental app purchasing behavior.
Underlying Intentions and Motivations of App Purchasing
The integrated model of behavioral prediction, which explains that behaviors are
ultimately influenced by attitudes, perceived behavioral control and perceived norms (Fishbein & Ajzen, 2010), was used as the basis for investigating what underlying factors influence parents to
purchase apps for their children. The model used in this study examined attitudes, perceived educational value, and perceived behavioral control as predictors of app purchasing behavior. App purchasing behavior was classified as moderate (purchasing up to 5 apps in the past 6 months) or frequent (purchasing more than 10 apps in the past 6 months). The model
successfully predicted whether parents were likely to buy up to 5 apps or more than 5 apps (i.e., whether a parent would be a moderate purchaser) and whether parents were likely to buy more than 10 apps (i.e., whether a parent would be frequent purchaser).
Interestingly, although many parents indicate that they buy apps because they believe them to have educational value, it was found that the belief that apps are educational and the belief that they can provide educational outcomes is not a predicting factor in parental app purchasing as hypothesized in H1. In terms of theoretical implications, this finding supports the existing notion that attitudes and perceived educational value are independent predictors in parents’ media decisions. While parents’ app purchasing behavior is greatly influenced by their general attitudes towards apps it is not influenced by the perceived educational value of apps. It is possible that parents believe that apps in general could potentially have educational value but do not fully understand what that value could be and whether they can adequately make that judgment when evaluating a specific app. Perhaps parents are not confident in their ability to judge whether or not apps may results in educational outcomes which is why they are influenced by their more general attitudes and not specifically by perceived educational value at the moment of purchase. Parents may be more confident in their ability to judge whether their children will enjoy using the app then in their ability to determine whether their will actually learn from the app. Faced with multiple apps that all imply educational results, it is possible that parents do not trust their own judgment to determine which app will definitely lead to the educational outcomes it describes and choose to rely on their more general attitudes towards apps that do not
necessarily concern the educational value of apps.
Parents’ general attitudes towards apps were found to be a predictor of purchasing apps for their children as hypothesized in H2. In fact, attitudes were found to be the strongest
predictor of this behavior. In line with previous research regarding parental attitudes and media (Barr et al., 2010; Vaala & Hornik, 2013), the more positive parents are about the medium, the more likely they are to expose their children to it. In this case, parents who hold positive attitudes
towards apps and believe they lead to beneficial outcomes for their children are more likely to purchase them. This finding has theoretical as well as practical implications. Academically, this finding sheds light on the behavioral process of app purchasing among parents and focuses the research lens with which to investigate this behavior. Researchers can now use the knowledge of attitudes’ influence on parents as a starting point. Research can extend this finding to study how these attitudes come to be and whether, for example, attitudes are influenced by age, gender, or educational background. In terms of practical implications, the notion that general attitudes are driving parents to purchase apps regardless of educational value may prove to be extremely valuable to app developers in the ways in which they develop and describe their apps. It may be possible for app developers and marketers to avoid all conflict regarding educational marketing claims by focusing more on appeals to parents’ general attitudes about apps that do not
necessarily revolve around educational outcomes as these two variables were proven to be only moderately correlated. Such appeals may include outcomes of fun and entertainment because it seems that while parents do indicate that they purchase apps because they believe them to be educational this belief alone is not driving purchasing behavior. It is parents’ more general attitudes that are at the core of app purchasing.
Parents’ perceived control over their ability to select apps that are appropriate for their children was not proven to predict app purchasing behavior as hypothesized in H4. While most parents did indicate a high sense of this type of perceived control, it does not seem to influence their app purchasing decisions indicating that these decisions are influenced by other factors (i.e., general attitudes and perceived control over app use). It is possible that parents may believe they can choose app content based on whether their child will enjoy using the app, but do not believe they have the ability to adequately judge whether the app’s content will be educational and lead to educational outcomes. Control over app selection was tested in this study in regards to parents’ perceived ability to select apps they believe were appropriate for their child and that their child would enjoy. This did not specifically measure whether parents believed they could chose appropriate educational content and whether they believed they could adequately choose apps that would lead to educational outcomes. Future research should investigate the notion that parents may believe apps can be educational in general but do not believe they have the power to sufficiently judge apps and determine their educational value.
Unlike perceived behavioral control over app selection, perceived behavioral control over app use was in fact found to have a relationship with parental app purchasing. Interestingly, the relationship found between the variables was negative. Although this finding does not support the hypothesis (H3) it does provide interesting and important implications. These findings suggest that the more control parents believe they have over how their children use apps (in terms of the time they use them and the content they are exposed to within them) the less likely parents are to purchase many apps for their children. A possible explanation for this finding is that parents that are very controlling of their children’s app use are not overly positive about apps or about media in general. If parents feel the need to excessively limit children’s use of apps and scrutinize the content the children are exposed to this might indicate a more negative attitude towards media and, therefore, less app purchasing. Most parents in the study indicated that they do in fact implement rules in their household to limit their children’s time on apps signifying that parents may be concerned about negative media effects.
Vaala and Hornik (2013) found that the less control mothers felt over their children’s television viewing the more television their children watched. It is possible that parents are actively enforcing their control over app use to avoid excessive use by their children. More research is needed to determine if control over app use does in fact stem from negative attitudes towards apps and/or media and vice versa.
The Influence of Educational Marketing Claims on App Purchasing
The third and final aim of this study was to extend the research done by Vaala and Lapierre, (2013) on educational marketing claims of children’s DVD’s to app marketing specifically. The goal of the experimental manipulation performed in this research was to
determine whether there is a difference in how parents evaluate apps based on their descriptions. These descriptions being either direct in the educational outcome they indicate or vague in such statements (e.g., a description that specifically states the learning outcome [your child will learn 100 words] as opposed to a vague outcome [your child will be exposed to new vocabulary words’]. In line with expectations based on previous research, findings suggest that there is no difference in perceived educational value (H5), attitudes (H6), or purchase intent (H7) following exposure to direct and vague marketing claims. Parents in the study formed similar attitudes towards the app presented to them, regardless of the wording in the descriptions. Furthermore,
parents rated the app in very much the same way in terms of how educational they believed it to be and indicated the same intent to purchase the apps.
Overall parents found the app with the vague description just as educational as the app with the direct description. The use of less direct words in the description did not change parents’ perception of the app’s educational value. Interestingly, analysis of the manipulation check of this study suggests that parents that were exposed to the vague marketing description, in
comparison the parents exposed to the direct marketing description, thought their children would learn less words and less numbers. Yet, there was no significant difference in how both groups evaluated the app’s overall educational value.
It is possible that parents exposed to the vague description believed that the words and numbers they thought their child would learn is sufficient and beneficial to the child meaning they were content with the educational value communicated through the vague description. Without the description indicating exactly how many words and numbers the app could teach (as done by the direct description) parents were left to guess how much their child would learn in each category based on the vague claims. Parents exposed to the vague description estimated on average that their child would learn 16 new words and 5 new numbers. Perhaps parents felt that any improvement in their child’s vocabulary and knowledge of numbers (even if not as large as the direct description indicated) would be valuable and would therefore perceive the app as having high educational value. Alternatively, it is possible that although these parents anticipated relatively lower figures of words and numbers learned from the app, they believed the app could have additional educational outcomes not stated in the description at all such as, for example, the development of hand-eye coordination. More research is needed to investigate exactly what educational outcomes parents believe apps can lead to and to what extent these outcomes are mentioned in descriptions as opposed to independently thought of by parents.
In general, all parents (i.e., those exposed to the vague description and those exposed to the direct description) formed very positive brand attitudes towards the app. Additionally, parents expressed the same intentions to buy both the app with the vague description and the app with the direct description. The app proved to be popular among parents with the majority of the sample indicating they would likely purchase it for their child. Overall, parents formed the same impressions regardless of the vague marketing claims. This is an important finding seeing as the
apparent solution to the ongoing debate regarding the allegedly misleading marketing of educational media was for companies to change their promotional statements as to not directly imply educational outcomes. This form of academic research could potentially be used to
demonstrate the perhaps new policies need to be considered as the change of marketing to vague claims does not yield significant differences in parents’ perceived educational value of the app, their attitudes towards the app, or their intention to buy the app.
Limitations and Future Research
It must be noted that, like most social science research, this study has limitations that should be considered when evaluating its findings. Firstly, the sample used in the research although sufficient in size was not necessarily representative of all parents. The majority of the parents in the study were highly educated and from modern countries. While it could be argued that these parents do in fact represent the parents that have access to devices such as smartphones and tablets (that would give them access to apps) future research should try to incorporate
parents from different layers of society. Such parents may not have the same access to the app store and may not download apps on a regular basis, leading to very different experiences and habits regarding app purchasing that should also be studied. Moreover, these parents might be somewhat less educated than the parents in this study which may influence the ways in which they evaluate, and are influenced, by marketing. Additionally, the sample used in this study was predominantly female which may have influenced results. It is possible that males and females evaluate apps differently and are influenced by different factors. Future research should
incorporate more male participants to assess whether there is a significant difference in the ways in which males and females purchase apps for their children.
Secondly, the experimental manipulation used in this study (i.e., the app description) was designed to mimic a generic app found in the app store however it is possible that other aspects of it (such as the app name, the character or the picture) influenced parents’ evaluation of the app. For example, it is possible that parents judged the app based on the relatively low quality of the image (as it was not designed by a professional app developer) or based on the name of the app that did not imply educational content. These factors may have influenced some parents to rate the app as less favorable, less educational, or lower their intent to purchase the app. Alternatively, it is possible that the image of the app (which portrayed numbers and letters)
influenced parents to give the app higher ratings in terms of its educational value regardless of the verbal description. Future research should expand this study’s findings by testing if and how other aspects of app descriptions (e.g., app name, app picture, and perceived quality of the app) influence parents in their purchasing decisions.
Thirdly, while this study did differentiate between moderate and frequent app purchasers it is important to note that only a small portion of the sample (29%) were in fact frequent
purchasers that have purchased more than 10 apps for their child in the past 6 months. Frequent purchasers may have more experience evaluating apps and app descriptions and should be studied further to evaluate differences among app purchasers. However, future research should consider the low percentage of parents that purchased more the 10 apps and reassess what should constitute a frequent purchaser. It may also prove to be beneficial to study these behaviors over longer periods of time (i.e., longer than a 6 month period) to assess how frequently parents buy apps and the intervals of purchases.
Lastly, seeing as purchasing behavior was tested in this study theoretically it does not indicate actual purchasing but rather purchasing intentions. If possible, future research should allow actual downloading/purchasing of the app in real time to provide a more reliable measure. Furthermore, purchase intention indicated in the study does not pertain to any specific price of the app. Future research should examine whether price strongly influences parents’ attitudes and purchasing intentions of the app as some apps in the app store are free while other vary in price. The ability to purchase apps for certain prices (i.e., having the financial ability to afford them) may be considered in future research as a form of perceived behavioral control.
Despite its limitations, this study makes substantial progress in understanding parents and their purchases of apps for their children. While this study does not support nor negate the notion that apps can in fact educate children, it does illuminate the existing difficulty in app marketing and provides input for the debate regarding this issue. There may be a need to reevaluate the policies regarding marketing of educational apps seeing as parents are evidently forming similar opinions and evaluations of apps with direct or vague descriptions. It is clear that more research is needed to shape and construct appropriate guidelines for app developers to adhere to in order to fairly market their products until they are able to sufficiently back-up their claims. Especially considering the fact that parents indicated in this study that they regularly read descriptions and
that these are important to them when they make app purchasing decisions. As apps grow more and more popular it is crucial to continue to study what influences parents’ app purchasing and how research and policy can help them make better-informed decisions.
References
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision
Processes, 50(2), 179-211.
Arvola, A., Vassallo, M., Dean, M., Lampila, P., Saba, A., Lähteenmäki, L., & Shepherd, R. (2008). Predicting intentions to purchase organic food: The role of affective and moral attitudes in the Theory of Planned Behavior. Appetite, 50(2), 443-454.DOI:
10.1016/j.appet.2007.09.010
Bachman, K., (2013). Advocacy group says mobile apps for toddlers aren’t very smart. Adweek. Retrieved from http://www.adweek.com/news/advertising-branding/advocacy-group-says-mobile-apps-toddlers-arent-very-smart-151744
Barr, R., Danziger, C., Hilliard, M. E., Andolina, C., & Ruskis, J. (2010). Amount, content and context of infant media exposure: a parental questionnaire and diary
analysis. International Journal of Early Years Education, 18(2), 107-122. DOI:10.1080/09669760.2010.494431
Brown, A. (2011). Media use by children younger than 2 years. Pediatrics, 128(5), 1040-1045. DOI: 10.1542/peds.2011-1753
Chiong, C. & Shuler, C. (2010). Learning: Is there an app for that? Investigations of young
children’s usage and learning with mobile devices and apps. New York: The Joan Ganz
Cooney Center at Sesame Workshop.
Courage, M. L., & Howe, M. L. (2010). To watch or not to watch: Infants and toddlers in a brave new electronic world. Developmental Review, 30(2), 101-115. DOI:
10.1016/j.dr.2010.03.002
DeLoache, J. S., & Chiong, C. (2009). Babies and baby media. American Behavioral
Scientist, 52(8), 1115-1135. DOI: 10.1177/0002764209331537
Dogtiev, A., (2014). Marketing your apps to kids? Here’s what parents think is important. App
