Although many teacher candidates use apps daily, understanding what an app is and having a context for how they have evolved will help build their conceptualizations of apps.\u00a0 An app is essentially a small computer program that can be quickly downloaded onto a mobile computing device, (such as a tablet or smartphone) and immediately engaged without rebooting the device (Lucey, 2012; Pilgrim, Bledsoe, & Reily, 2012). As of October 2013, most apps were programmed to run on eitherApple\u2019s iOS operating system (Apple Inc., 2013) or Android, Google\u2019s Linux-based operating system (Droid, 2012); consequently, Apple and Google are the world\u2019s largest providers of apps (Godwin-Jones, 2011).<\/p>\n
There are both free apps that do not require money to download and paid apps that cost users $0.99 or more to download.\u00a0 Founded in 2008 with 500 apps, Apple customers are directed to the App Store to browse, purchase (if necessary), and download apps for their iPads, iPods, and iPhones.\u00a0 As of May 2013, Apple customers had downloaded more than 50 billion apps (Apple Press Info, 2013; Woollaston, 2013).<\/p>\n
Also founded in 2008 and formerly known as the Android Market, Google Play allows Google customers to browse, purchase (if necessary), and download apps for their Android smartphones and tablets.\u00a0 By June 2012, 20 billion apps had been downloaded from Google Play (Fingas, 2012).<\/p>\n
As of October 2013, App Trace (http:\/\/www.apptrace.com<\/a>), a website specializing in app analytics, reported that 618,064 free apps and 382,867 paid apps were available to be downloaded.\u00a0 However, teachers searching the App Store or Google Play will likely encounter challenges when identifying apps to use with their students based on their app classification system.<\/p>\n Challenges in Selecting Apps by Subject Area and Function<\/strong><\/p>\n With over 20,000 educational apps available for download (Rao, 2012), teachers visiting the App Store or Google Play can quickly become overwhelmed when seeking to select the most effective app for their instructional needs (McGarth, 2013; Walker, 2011). In the App Store, for instance, when teachers search for \u201ceducation\u201d apps, they are presented the menu shown in Figure 1.<\/p>\n <\/p>\n From these menu options, teachers must decide how they want to browse apps.\u00a0 If they choose to browse the menus for \u201cBest New Apps\u201d or view apps by grade level (e.g., apps for Middle & High School or apps for Preschool & Elementary) on this screen, the App Store lumps apps together using no specific organization pattern.\u00a0 This lack of organization results in teachers having to search through apps that could be used in any subject area and for a variety of purposes.<\/p>\n The first app shown for Middle & High School (Figure 1) is Notability, an app used to help students take notes, followed by Math 42, an app designed to assist students in solving math problems, and Starting Shakespeare, an app designed to introduce students to Shakespearean works.\u00a0 The lack of organization impedes teachers from finding apps quickly and efficiently because the App Store clusters apps together without considering subject area implications.<\/p>\n A middle school social studies teacher looking for an app using this menu would have to scroll to the 18th app before finding Today\u2019s Documents by the National Archives (not shown in Figure 1), the first app indexed by the App Store that is specific to the social studies classroom.<\/p>\n Today\u2019s Documents shows a primary source and provides background about how the primary resource is connected to a specific calendar.\u00a0 If that social studies teacher wanted to use an app to develop students\u2019 abilities for analyzing primary sources, this app would be beneficial.\u00a0 Yet, to find an app to develop students\u2019 map reading skills, the teacher would have to scroll to the National Geographic World Atlas app, which is the 42nd app listed in this index, to find an app that includes maps.<\/p>\n Because the App Store does not use an effective organizational strategy for categorizing apps on its main menu for education apps, teachers may spend a significant amount of time trying to locate apps that align with their instructional needs.\u00a0 However, if teachers select the \u201cEducation Collections\u201d option from the main menu, the App Store does a better job in categorizing apps.<\/p>\n When selecting an Education Collections option such as Apps for Every Grade, Apps for Middle School, or Apps for High School, the App Store loads a menu that allows teachers to select a specific subject area, as shown in Figure 2.<\/p>\n Figure 2. <\/strong>Screenshots of the App Store\u2019s menu that allows high school teachers to select a specific subject area, March 31, 2014.<\/em><\/p>\n <\/p>\n The options provided in this menu, however, are not consistent.\u00a0 For example, whereas the subdisciplines for Math (e.g., Algebra, Geometry, and Calculus) and History (e.g., American History, World History, Civics, American Government, and Economics) are lumped together, the subdisciplines for Science (e.g., Biology, Chemistry, Physics, and Earth & Space) are listed as separate subject areas.\u00a0 Yet, English language arts is not classified as a subject area; rather, it is renamed Literacy.\u00a0 Teachers of any discipline trying to locate apps must select the option that most closely aligns with their subject area, which again becomes complicated when considering apps\u2019 purposes.<\/p>\n Therefore, if biology teachers are looking for an app, they would select the Biology option from the menu shown in Figure 2, and they would be presented with the classifications shown in Figure 3.<\/p>\n <\/p>\n The challenge is that even these more nuanced classifications do not consider the function of specific apps, and the apps classified as Plants & Animals illustrate this challenge.<\/p>\n For biology teachers searching for apps about endangered species, they would likely be drawn to the apps classified as Plants & Animals on this screen.\u00a0 When browsing these apps, biology teachers are presented with apps that index animals, include videos of animals in their habitats, and guide students through virtual animal dissections.\u00a0 Because each of these apps has a distinct function\u2014such as presenting information to students in the form of text, images, and videos; quizzing students about related topics; and letting students explore animal cadavers\u2014biology teachers must review these apps until they find one that meets their specific needs, which can become complicated.\u00a0 Biology teachers who want their students to complete a research report about an endangered species would need to choose multiple apps.\u00a0 Specific apps needed for this assignment include apps that present information to students, ask students questions about endangered species, and let students document their learning in the form of a document, presentation, song, video, or image.<\/p>\n Teachers likely could identify apps that present information about endangered species and perhaps find an app that asks students questions about endangered species; however, they would not find an app that allows students to document their learning.\u00a0 To find that type of app, teachers would need to browse the entire App Store\u2019s Education section, because there is no specific category for apps with that type of functionality when searching for apps by subject area.\u00a0 This lack of clarity when considering how apps are classified by subject area and function carries over to the website databases that index apps.<\/p>\n Existing Databases for Selecting Educational Apps<\/strong><\/p>\n Outside of the App Store, Google Play, top apps for education lists, and app recommendations from colleagues and leaders of professional development sessions (Highfield & Goodwin, 2013), only a handful of databases are designed for parents and teachers to search for educational apps.\u00a0 Examples of these databases include Graphite.org<\/a> created by the Common Sense Media company, Apps4Edu<\/a> created by the Utah Education Alliance, EdShelf<\/a> created by the EdShelf company, and Appitic <\/a>supported by Apple Distinguished Educators.\u00a0 [Editor\u2019s Note:<\/em><\/strong> Website URLs can be found in the Resources<\/a> section at the end of this paper.]<\/p>\n Each of these databases contains searchable indexes of educational apps that provide users with a description, general evaluation, subject area classification, and information about the app\u2019s cost and device compatibility.\u00a0 Appendix A<\/a> shows screenshots related to how these databases provide information to their users.\u00a0 Although these considerations are useful in a general sense, they are of limited use to the teacher education community, because an app\u2019s purpose is not included in the information these databases provide.<\/p>\n Illustrating the Significance of Purpose<\/strong><\/p>\n Whereas these website databases allow teachers to search for apps using more specific criteria than does the App Store, they still provide only limited data about specific apps that can be grouped together.\u00a0 Although multiple apps may seem to address the same subject area, skill, content, or knowledge, the way they do so differs, resulting in widely varying effectiveness.\u00a0 For example, both the apps MeMe Tales and One Minute Reader are designed to develop young children\u2019s reading abilities, but they use very different approaches.\u00a0 MeMe Tales is designed to build enjoyment for reading by allowing children to choose a colorful book and then decide if they would like to read the book aloud or have the app read it to them.<\/p>\n After reading the book or having it read to them, MeMe Tales awards children points (Figure 4).\u00a0 In this way, MeMe Tales encourages children to enjoy reading without using assessments.<\/p>\n <\/p>\n On the other hand, One Minute Reader is designed to develop children\u2019s reading skills by using a variety of fluency and comprehension assessments.\u00a0 One Minute Reader first assesses the amount of words children can read per minute by timing them on a cold (first time) reading passage. Next, the app models a fluent reading of the same passage by reading it aloud.\u00a0 Children then read the same passage a third time (hot reading), and the app tracks the number of words read per minute to determine if there was an increase in reading fluency during the third reading as compared to the cold reading score (Figure 5).<\/p>\n Review of Existing App Classifications Systems <\/strong><\/p>\n Although the advent of iPads and tablets is fairly recent, the presence of computers in classrooms is not.\u00a0 Likewise, evaluation of educational software is nothing new. Researchers have been studying how computers impact student learning for decades (Clements & Nastasi, 1988; Haugland, 1992; Wolfram, 1984).\u00a0 Therefore, it is logical to begin an evaluation of instructional apps by borrowing from extant practice, then refine as needed.\u00a0 This strategy provides insight into how researchers classified computer software, as well as to demonstrate how the task of creating a classification framework for categorizing instructional apps builds upon the work of previous researchers.\u00a0 Because the evaluation process largely differs from the classification process, we focused specifically on the selection of articles representing how researchers established organizational frameworks in classifying educational software, not assessing the quality of educational software.<\/p>\n In an early analysis of how schools were using computers after they were first introduced, Pelgrum and Plomp (1993) used a questionnaire to collect data from 60,000 educators worldwide.\u00a0 As part of the questionnaire, respondents were asked to classify their school\u2019s available educational software into one of 23 categories (see Appendix B <\/a> for a complete list of the categories).\u00a0 Although Pelgrum and Plomp did not provide a description for the categories, they did identify that so-called drill-and-kill software, educational games, and word processing programs were the three most frequent uses of computers in elementary schools.\u00a0 Conversely, the software identified as least used in elementary schools included programs for interfacing with labs, producing interactive videos, and designing\/making products.<\/p>\n Additionally, when comparing software used in elementary schools to software used in lower secondary (middle) and upper secondary (high) schools, Pelgrum and Plomp found that word processing programs, software that taught programming language, and spreadsheet and database programs were the four most frequently used.\u00a0 Similar to elementary school findings, the software used least frequently in lower and upper secondary schools included programs for interfacing with labs, storing questions for tests, and producing interactive videos.\u00a0 A critique of Pelgrum and Plomp\u2019s work is that their classification framework considered only the function of the software instead of what users are able to do with the software (McDougall & Squires, 1995).\u00a0 As such, other classification systems for categorizing educational software were later developed.<\/p>\n Bruce and Levin (1997) proposed a taxonomy for classifying educational technologies that consisted of four main categories with subcategories embedded in each.\u00a0 Bruce and Levin chose the term media <\/em>instead of software or programs \u201cto shift the focus from the features of hardware or software per se to the user or learner\u201d (p. 83).\u00a0 In this way, they honored McDougal and Squires\u2019 (1995) suggestion by considering how users engage the technology being classified.<\/p>\n The first category of Bruce and Levin\u2019s taxonomy is Media for Inquiry, which contains four subcategories: (a) Theory Building, (b) Data Access, (c) Data Collection, and (d) Data Analysis. Each subcategory is designed to support users in formulating and answering questions.\u00a0 For example, when users are working to answer a question, they would use the Theory Building subcategory to assist them in understanding the question. \u00a0Next, they would use the Data Access subcategory to identify previously discovered knowledge about that question.\u00a0 Users would then draw on the technologies in the Data Collection subcategory to assist them in gathering and storing information related to their work.\u00a0 Finally, the Data Analysis subcategory would support users in interpreting the data collected.\u00a0 Thus, the Media for Inquiry technologies are classified together because they all support users as they seek knowledge to answer their inquiries.<\/p>\n The category Media for Communication contains four subcategories: (a) Document Preparation, (b) Communication, (c) Collaborative Media, and (d) Teaching Media. Each subcategory is designed to support users in sharing their work.\u00a0 For example, Document Preparation contains technologies to assist users in creating artifacts of their work, then users engage the subcategories of Communication technologies to share those artifacts.<\/p>\n The Collaborative Media subcategory consists of technologies that allow multiple users to contribute to one artifact together, and the Teaching Media subcategory contains technologies to support users in teaching the knowledge they produced or gained when using the Media for Inquiry technologies.\u00a0 The technologies in the Media for Communication category were classified together because they all help users disseminate information related to their inquiries.<\/p>\n The third and fourth categories do not contain subcategories, and Bruce and Levin (1997) described each of these categories with bulleted lists.\u00a0 For instance, the category Media for Construction is designed to include technologies that help users \u201caffect the physical world\u201d (p. 6).\u00a0 This category would include technologies that allow users to design pieces of architecture or alter physical landscapes.\u00a0 Media for Expression, their final category, includes technologies for creating images, audio recordings, and multimedia projects.\u00a0 The major difference between these two categories is that technologies in the Media for Construction category result in concrete, tangible objects (e.g., blueprints for a house), while technologies in the Media for Expression category result in digital artifacts (e.g., a slideshow movie or a music video).<\/p>\n In their review of technologies, Bruce and Levin found that 58.9% were aligned to the Inquiry category, 36.9% were aligned to the Communication category, 4% were aligned to the Construction category, and no technologies applied to the Expression category.\u00a0 Bruce and Levin claimed that technologies were not included in the Expression category because \u201cpersonal expression in the sense Dewey meant\u201d (p. 90) was not emphasized, and they recommended additional research be conducted in response to this finding.<\/p>\n Although multimedia composition software and animation programs that allow users to personalize their projects are now available, these technologies were still developing at the time of Bruce and Levin\u2019s work, which may also explain why their Expression category did not include any technologies.<\/p>\n After analyzing these two types of classification frameworks\u2014a framework that focuses only on the functionality of the computer software and another that considers how users engage the computer software\u2014we determined that a classification system must consider both the purpose and practice of the technologies being classified, and the use of both categories and subcategories was an effective grouping strategy.\u00a0 With this in mind, we then searched for classification frameworks constructed specifically for apps.<\/p>\n Searching for preexisting classification frameworks was challenging because the field of educational apps is still a relatively new topic of research.\u00a0 After experimenting with different search terms that used varying word combinations of app<\/em>, education<\/em>, and classify<\/em>, the search term \u201ceducation apps\u201d classification <\/em>was selected.\u00a0 This search term was selected because when other terms were used, nearly all the articles reported were off topic or did not relate to classifying educational apps.\u00a0 Consequently, when \u201ceducation apps\u201d classification <\/em>was used, two studies were found.<\/p>\n Handal, El-Khoury, Campbell, and Cavanagh (2013) reviewed more than 100 apps designed for primary and secondary math classrooms.\u00a0 The authors classified the apps into one of nine categories by analyzing each app\u2019s content for \u201cthe kind of learning activities associated with the app, the instructional experiences supported by the app and their media richness, [and] the anticipated levels of cognitive involvement and users\u2019 control over their learning\u201d (p. 144).<\/p>\n While this article did speak to McDougal and Squires\u2019 (1995) demand to consider the learner\u2019s interaction with the software, but the classification system\u2019s limitation only to math apps made it challenging to extend their findings to the other subject areas.\u00a0 Additionally, the nine categories the authors created convoluted their findings more than they helped to clarify.\u00a0 Having to consider nine different possible categorizations may overwhelm teachers when deciding how to use an app, which can consume teachers\u2019 limited instructional planning time.\u00a0 Therefore, we found the nine categories limited to math as being an inefficient app classification model.<\/p>\n Australian researchers Goodwin and Highfield (2012a,b) investigated the pedagogical design of 240 apps, for which they created three classifications.\u00a0 Their classifications included Instructive, Manipulable, and Constructive apps (and two hybrid classifications: Constructive\/Manipulable and Manipulable\/Instructive).<\/p>\n Instructive apps are described as promoting rote memorization of content through drill-and-skill activities.\u00a0 These apps require students to practice a skill repeatedly in order to increase their accuracy using the skill, and these apps can be likened to digital worksheets. \u00a0Manipulable apps provide students with guided discovery, which allows them to make choices about the topic they are learning and how they demonstrate that learning using a preconstructed context, template, or structure.\u00a0 For example, students using a search engine to gather information about a topic would be using a Manipulable app because they have the freedom to select their search terms, but they are limited to the results reported by the search engine.<\/p>\n Constructive apps are described as providing students with open-ended contexts, templates, or structures allowing them to create learning artifacts (e.g., images, videos, or texts).\u00a0 For instance, when students create a PowerPoint presentation, they engage an open-ended template, in which they have the power to design the presentation and its content to their liking.<\/p>\n Of the 240 apps they reviewed, Goodwin and Highfield labeled 75% as Instructive, 23% as Manipulable, and only 2% as Constructive.\u00a0 They concluded that an alarmingly high number of apps targeted for young learners are Instructive, and they called for more Constructive apps to be developed.\u00a0 In reviewing their research, we supported Highfield and Goodwin\u2019s concepts for delineating apps by their instructive, manipulable, and constructive qualities; however, they did not fully develop these categories.<\/p>\n While Highfield and Goodwin\u2019s (2012a,b) framework provides a structure for classifying educational apps, it does not consider the value of an app\u2019s specific skills or purposes.\u00a0 Rather, they made generalizations about large groups of apps. \u00a0For example, Tap Quiz Map is an app that uses memorization strategies and quizzes to help students recall specific geographical places, and its purpose is for students to remember the names and locations of exact places.\u00a0 Another app, Phonics Genius, repeatedly shows and pronounces the sounds of letter combinations to build students\u2019 phonological awareness, and its purpose is to develop students\u2019 reading abilities.<\/p>\n Using Highfield and Goodwin\u2019s classification system, both of these apps would be labeled as Instructive, because they rely on memorization to teach their content.\u00a0 The problem is that the value of the skill being taught is not considered when using their classifications; instead, Highfield and Goodwin consider only the instructional method. Whereas students can look up the capital of Oklahoma in order to answer a Tap Quiz Map question, they would not be able to read that information without being phonologically aware, which is a skill that Phonics Genius develops.<\/p>\n By labeling all apps that use repetition to build students\u2019 knowledge as instructive, Highfield and Goodwin are discounting that some instructive apps teach students relevant, foundational skills needed to engage manipulable and constructive apps, while other apps, such as Tap Quiz, teach only isolated knowledge that has little use to students (Daggett, 2005).\u00a0 To prepare teacher candidates to integrate apps effectively into their instructional practices, a classification system is needed that uses categories and subcategories to analyze the skills apps teach, recognizes how apps are interrelated, and provides examples of these skills and interrelations.<\/p>\n Grouping apps into disparate categories without also considering the purpose of the knowledge students acquire from the apps does not help teacher candidates make these vital connections.\u00a0 However, a framework that breaks Highfield and Goodwin\u2019s classification system down into more precise categories and subcategories and is more closely aligned to an established framework of teacher knowledge, such as technology, pedagogy, and content knowledge (TPACK; Koehler & Mishra, 2005, 2009), would recognize these subtle nuances.<\/p>\n Methodology for Developing a Framework for Categorizing Apps<\/strong><\/p>\n When planning how to conduct this work, we used qualitative research methods (Coffey & Atkinson, 1996; Glense, 2006), because classifying apps is an act of interpretation, which requires the researcher to become a research tool (Connelly & Clandinin, 1990). We also selected apps across the subject areas of English language arts, mathematics, science, and social studies.\u00a0 Unlike the classification system created by Handal et al. (2012) that classified apps only for the mathematics classroom, our framework was closer to the one designed by Goodwin and Highfield (2012a,b), which can be used by all subject areas teachers.\u00a0 Although apps for art, music, and physical education among several other subject areas exist, it was more practical to limit our framework to the core curriculum subject areas present in public schooling.<\/p>\n We agreed that practicality and ease of use were important components of our work, and we found that the educational sections of both the App Store and Google Play contain apps for each subject area.\u00a0 However, we still needed to reduce the number of apps we were to review for our classification system to a practical number using specific selection criteria.<\/p>\n After working with these initial criteria, we discovered that not all educational apps have a specific subject area.\u00a0 For example, educational apps such as Educreation and Popplet can be used in multiple subject areas.\u00a0 Educreation is an app that allows users to create a shareable video about a specific topic.\u00a0 Popplet allows users to create conceptual maps and graphic organizers about most any topic.<\/p>\n Both apps can be used after reading a specific content-area text.\u00a0 For example, students can respond to the text they read by creating a video using the Educreations app to elaborate on the text or offer a critique of it.\u00a0 Students can use the Popplet app to design a conceptual map in response to the text to display their understanding of it.\u00a0 Because both of those apps can be applied to multiple subject areas, the selection of apps by subject area only was impractical. \u00a0In response, we identified non-subject-specific apps by reading the description of the app posted in the App Store, which led us to the final filter.<\/p>\n Two other relevant types of apps are Instructional and Teacher Resource.\u00a0 We defined Instructional apps as apps that help teachers deliver content knowledge or build students\u2019 critical thinking skills and Teacher Resource apps as apps that help teachers manage a class or plan lessons.\u00a0 Teacher Resource apps were not included in our framework.\u00a0 For instance, Class Dojo is an app designed to support teachers in managing their classrooms, and we labeled it as a Teacher Resource app because it does not help teachers deliver instruction to students.\u00a0 The Common Cores State Standards app is also a Teacher Resource app because it helps teachers plan instruction, not deliver it.\u00a0 As a result, we delimited our framework to Instructional apps.\u00a0 We used this selection criterion because we wanted to work with a pool of apps to which all teachers had access.<\/p>\n Selecting Apps<\/p>\n To select apps, we consulted the App Store for English language arts, math, science, and social studies instructional resources from across the preK-12 spectrum.\u00a0 We searched the App Store using the terms English language arts<\/em>, science school<\/em>, social studies<\/em>, and math school <\/em>(the word school<\/em> was added to science <\/em>and math, <\/em>because those two terms are widely used outside of educational contexts; adding the word school to each term narrowed our search to apps designed for educational purposes).<\/p>\n Although some apps can be applied to multiple subject areas (e.g., Educreation and Popplet), several apps can be applied to one specific subject area.\u00a0 To establish validity (Messick, 1989), we chose 10 apps for each subject area that met our selection criteria.\u00a0 Ten apps provided a large enough sample size within each category to gain an understanding of the types and functions of apps designed for the subject area.\u00a0 Additionally, many apps have been created to support students\u2019 standardized test readiness.\u00a0 As such, eight apps designed for this purpose were included.\u00a0 Furthermore, when cross-referencing Shuler\u2019s (2009) review of the top 100 educational apps based on age and subject area, we that found apps for teaching literacy, general learning, and math to toddlers and preschool students represented a significant amount of the most popular apps downloaded.\u00a0 Therefore, we purposely selected 10 apps that targeted building toddlers\u2019 and preschoolers\u2019 early literacy skills and 10 apps that targeted building their early numeracy skills.<\/p>\n Finally, because the apps we included at this point were subject-specific apps, test-readiness apps, or apps targeting young learners, we wanted apps that could be used by multiple content areas and be used to create learning artifacts.\u00a0 As such, we consulted Appitic because it allows users to search its database according to Bloom\u2019s Taxonomy, not only by subject area.\u00a0 Using Appitic enabled us to identify 15 apps that could be applied across subject areas.<\/p>\n Last, because teachers may consult blogs and online lists of best apps for education, we conducted an Internet search using Google and the search term \u201ctop apps for education\u201d to identify top online resources (Heick, 2013; Noonoo, 2013; Sawers, 2012 ) where teachers may go to select apps, excluding the databases we had already consulted. From these resources, we identified an additional nine apps, which brought our count to 92.By following this selection process, we minimized selection bias while generating a pool of diverse, (relatively) popular, and high-quality apps for analysis.<\/p>\n Categorizing Apps<\/p>\n Because analyzing apps involves understanding their functions and purposes, we followed a qualitative coding methodology (Coffey & Atkinson, 1996; Glense, 2006).\u00a0 To code, we used Mean\u2019s (1994) categories for technology analysis\u2014tutor, explore, tool, and communicate\u2014as a basis for our framework, and we modeled Murray and Olcese\u2019s (2011) methodology for analyzing apps according to those categories.\u00a0 In this way, Mean\u2019s categories offered us a guide for thinking about how apps could be classified.\u00a0 For each app we identified, we first downloaded and used it to gain a general familiarity for it.\u00a0 Then we asked the following questions:<\/p>\n To ensure reliability, (Coffey & Atkinson, 1996; Merriam, 2009), two of this project\u2019s authors examined each app, and recorded responses using a spreadsheet.\u00a0 After all the apps were examined, commonalities related to their purpose, requirements, and usages were used to group the apps into skill-based, content-based, and function-based categories.<\/p>\n Also, a Pearson correlation coefficient was calculated for each category (Benesty, Chen, Huang, & Cohen, 2009). We found a .85 correlation for the skill-based category, a .85 correlation for the content-based category, and a correlation of 1.0 for the function-based category.\u00a0 As the numbers of apps placed in each category swelled and nuanced distinctions between the apps in the same categories were recognized, we collaborated to break the larger categories into subcategories.\u00a0 The resulting classification framework is as follows:<\/p>\n For representative apps in each subcategory, see Table 1.<\/p>\n Table 1 \n![\"Figure](\"https:\/\/citejournal.org\/wp-content\/uploads\/2016\/04\/v14i2General1Fig1.jpg\")
Figure 1. <\/strong>Screenshots of the App Store\u2019s Education section, March 31, 2014.<\/em><\/p>\n![\"Figure](\"https:\/\/citejournal.org\/wp-content\/uploads\/2016\/04\/v14i2General1Fig2.jpg\")
<\/p>\n![\"Figure](\"https:\/\/citejournal.org\/wp-content\/uploads\/2016\/04\/v14i2General1Fig3.jpg\")
Figure 3.<\/strong> Screenshots of the App Store\u2019s categories for Biology, March 31, 2014.<\/em><\/p>\n![\"Figure](\"https:\/\/citejournal.org\/wp-content\/uploads\/2016\/04\/v14i2General1Fig4.jpg\")
Figure 4. <\/strong>A screenshot of how MeMe Tales uses points as a reward for reading, March 30, 2014.<\/p>\n![\"Figure](\"https:\/\/citejournal.org\/wp-content\/uploads\/2016\/04\/v14i2General1Fig5.jpg\")
Figure 5.<\/strong> A screenshot of One Minute Reader reporting the amount of words read during a cold reading against the amount of words read during hot reading, March 30, 2014.<\/em>
\nLast, the app asks comprehension questions to monitor children\u2019s understanding of the passage (Figure 6).<\/p>\n![\"Figure](\"https:\/\/citejournal.org\/wp-content\/uploads\/2016\/04\/v14i2General1Fig6.jpg\")
Figure 6.<\/strong> A screenshot of a comprehension question used by One Minute Reader, March 30, 2014.<\/em>
\nMeMe Tales uses a low stakes, nonevaluative approach to developing children as readers, whereas One Minute Reader uses a higher stakes approach that features multiple assessment points.\u00a0 Because both MeMe Tales and One Minute Reader focus on developing literacy skills, existing app databases classify these apps only by subject area (e.g., Reading or Language Arts), which is misleading and counterproductive.\u00a0 By labeling them as Reading or Language Arts apps, the databases do not differentiate and report their specific purposes.\u00a0 The teacher must then conduct additional research to understand how the app functions and what its purpose is.\u00a0 A framework to support the classifications of apps that uses criteria for discerning the differences between apps with similar subject area implications\u2014in this case, developing children\u2019s reading abilities\u2014would help teachers consider which apps are most appropriate to use with their students.<\/p>\n
\nTo develop our framework for classifying apps, we reviewed 92 apps; Appendix C<\/a> shows a visual representation of our app selection and classification process.\u00a0 Because over 20,000 educational apps have been developed, purposeful sampling techniques were employed (Glense, 2006).\u00a0 Our first selection criterion was that the app had to be free of charge.\u00a0 Because policies about school districts and teachers being able to purchase apps vary greatly, we wanted to safeguard against any potential challenges related to cost.\u00a0 Using only free apps in developing our framework allowed us to avoid those potential pitfalls.<\/p>\n\n
\n
\n
\n
\n
\n<\/strong>Representative Apps for the Classification Framework\u2019s Subcategories\u00a0\u00a0\u00a0<\/strong><\/p>\n