Mobile data collection

Synonyms:
MDC, mobile phone logging, mobile technology
04_19_22---Motorola-Mobile-Phone_web.jpg

Mobile Data Collection (MDC) is the use of mobile phones, tablets or personal digital assistants (PDAs) for programming or data collection.

MDC can be very useful to the evaluator who is collecting quantitative data for their evaluation or abstracting data for an evaluation. 

There are many mobile phone applications (referred to as platforms) that will allow you to build a mobile data collection survey. These platforms will allow you to customise the survey to collect specific data as required, such as photographs, information from a list selection, voice recordings, GPS coordinates, etc. Platforms vary in ease of use, cost, and features.

    Examples

    Pact in Namibia wanted to conduct a retrospective evaluation of capacity development work with partners during their project. They needed to collect data from 17 partners, Pact staff themselves and the donor. Data collection had to be accomplished within a one week period.

    Mobile technology facilitated the collection of the evaluation data by ensuring that a survey with skip patterns and variable checks was designed and that the data were directly uploaded to the platform server by the evaluators. The data were available the next week for the analysis and report writing team, a process which aided the creation of a stakeholders presentation within one week of the end of data collection. The team learned a number of key lessons from this, and another use of mobile technology in Tanzania, which were presented at the American Evaluation Association meeting in 2012.

    Advice for choosing this method

    An important first step is figuring out whether or not MDC is appropriate for your evaluation. Most evaluations use surveys, which MDC is well-suited for and generally, a subset of available platforms cater best to creating surveys (as opposed to crowd-sourcing information, sending information out, or providing information on demand).  If you are collecting a lot of quantitative information and very little qualitative information, MDC is likely to work well.  However, if you are collecting many long, open-ended responses, MDC is unlikely to be the most efficient method; you could still record answers on mobile phones or tablets and either transcribe answers manually or use voice to text software, but typing in long answers on phones is likely to be disruptive to the interview process.

    • Consider your data needs. MDC is still not the best method for qualitative evaluation, but advances in this field are rapid, so this may change. 

    • MDC makes managing the data collection process easier.  Data are submitted in real-time, allowing managers to see the pace of data collection, the coverage, and which data collectors are submitting data when and from where.

    • MDC makes entering and aggregating data easier.  Because the data are entered into the main database at the same time as they are collected, the long process of transcribing and double-entering responses is eliminated.  Data can be analyzed as soon as the collection is done.

    • MDC offers instant visualization of data, including maps (if GPS coordinates are taken) and basic breakdowns of answers to each question.  This can be used for managing data collection or to aid later analysis.

    • MDC can lead to cleaner data.  Because platforms allow the form designer to put limits and skip logic on questions, answers that don’t make sense can be disallowed.

    • MDC can be less expensive than traditional paper surveys because it reduces the costs associated with printing and form transportation, double-entry, and data cleaning.

    • MDC allows the easy capture of other forms of data, such as images, video, and GPS coordinates.

    • What is the network coverage in the area where you’ll be collecting data?  While you don’t need network coverage to collect data, lack of coverage could affect when your data become available online.  If you don’t have network coverage, you can upload data through an Internet link, and you may want to invest in a portable WiFi hub.  Also, a few platforms will not work if there is no network coverage. 

    • What power sources are available to evaluators?  Will phones need to have a long battery life? Will you be able to recharge phones in vehicles?

    • What is your policy on phone use, ownership and replacement? Are data collectors allowed to use phones for personal purposes? Who pays for airtime? What happens if a phone is lost or stolen?

    • Consider the lifespan of the technology that you invest in. Will you be using the phones only once for a data collection effort, or will you repurpose the phones for other data collection or programming and use them throughout the life of the project? If you plan to re-use the phones, take this into consideration; you may want to invest more in phones that won’t be obsolete in a year, or ensure that they have certain capabilities that the survey may not require.

    • Ensure you have stakeholder buy-in for MDC. Getting buy-in from project stakeholders for the use of MDC is sometimes challenging. It’s important to clearly see and articulate the advantages of MDC, and to understand common challenges and solutions to those challenges so that they can be realistically addressed—and to understand when MDC might not be advantageous.

    • Be aware of any security risks of sending data. If you send data through a data network, the platform encryption will be very strong and secure (256-bit). This is not true of data sent by SMS.

    Advice for using this method

    Once you’ve determined to use MDC, you’ll need to go through the usual steps of evaluation design and implementation, but with some special considerations added.

    • Create the evaluation tool: Your evaluation tool will have a big impact on everything else you choose. Once you have a tool in place, you’ll know what characteristics to look for in phones and platforms.

    • Evaluate the platforms: Which has the features that you need for your survey? Which are most cost-effective?

    • Choose your technology: Which phones or tablets are compatible with the platform you’ve chosen? Which features (such as GPS capabilities, taking pictures or video, keyboards) do you need? What can the evaluation afford? What is locally available?

    • Create a budget: This should take into account the usual evaluation costs, such as data collectors, training workshops, and travel costs, but also include the cost of phones, airtime, the platform you’re using, and associated technology such as phone or solar chargers. The budget will not require paper copies and transport or double data entry, which sometimes means that the overall budget of mobile phone data collection will be lower.

    • Set up the mobile technology: Depending on the number of phones or tablets you are setting up, it may be easiest to set all of them up in advance for the data collectors, or to have the data collectors set up their own phones with the platform during the workshop.

    • Conduct the training: You will need an additional half to full training day to incorporate mobile technology training into your orientation for data collectors, depending on their level of familiarity with the technology.

    • Manage data collection: Data collection can be monitored in real time from a central online portal, making it easier to see if there are any problems with the questionnaire, if the survey is taking longer than planned, or if data collectors are covering their assigned areas properly.

    • Data aggregation: Your data are already aggregated. There is no need to re-enter the data, and you can export the data into Excel or a CSV format. All platforms will also provide some very basic instantaneous visualisation of data, which you can customise as necessary.

    • Data cleaning: Generally speaking, data will be much cleaner with mobile data entry than with transcribing paper forms, but data should still be checked over and cleaned.

    • Have backup options available: It’s wise for data collectors to have a couple of paper copies of the form when they go out into the field. The paper version may be easier for them to read questions off of and facilitates keeping their place in the survey, plus is a good backup in case the phone battery dies or something else goes wrong.

    Considerations when choosing a platform

    The platforms have very different characteristics, and there is currently no one platform that covers every base. The list below outlines common points of departure for phones and where the platforms tend to vary. However, note that all platforms are changing rapidly and that most are very willing to incorporate user feedback, especially if an organisation is planning on using the platform heavily; you may be able to negotiate different prices or services, or to suggest changes.

    Common points of departure for phones and where the platforms tend to vary

    Phone compatibility for forms: Nearly all platforms work with Android phones, but few work with Apple, Blackberry, and Windows devices.  Most will also work with Java phones that use apps.

    Skip logic: The majority of platforms offer skip logic, though not all.  Even among platforms with skip logic, platforms vary in ease of use; some platforms require coding or complex logic branches, others have quite simple mechanisms for both branch and prerequisite logic.

    Setup: Some platforms have setup support available for a fee.  Others may have free support contacts that can help you troubleshoot but won’t set up your survey for you.  Typically, platforms without survey setup support will be more user-friendly to set up yourself, while those with support will be more difficult to do yourself but may be more flexible in their capabilities.

    Language: Most platforms will let you ask questions in whatever language you have installed on your computer/phone, but some are coded in a way that makes this difficult and only offer certain alphabets. Additionally, some platforms are available in common languages such as Spanish, French, and Kiswahili.

    SMS: Platforms differ in whether or not they accept and send SMS.

    Web portal entry: A useful feature is being able to enter data through a web portal rather than through a phone, but not all platforms provide this.

    Extras: Check whether your platform can collect images, GPS, or video.

    While comparing platforms, you may also find that there are seemingly unimportant intangibles about a certain platform that make it more friendly to your intended users.  For example, one platform might prompt the data collector to move forward with a “next” button, while another may require that the question swipe to move forward; some data collectors may find one method more intuitive than another.

    Costs of the different platforms vary widely.  What is most cost-effective will depend on the nature of your data collection.  Pricing models include:

    1. Completely open source
    2. Freemium: a basic service is freely available, but extras such as SMS or technical support set up require either a subscription (SMS) or a one-time fee (for setup) which is usually somewhat steep
    3. Pay as you go: you pay for the amount of data collected
    4. Subscription: after a trial period, you pay a monthly or yearly fee to use the service

    Most platforms will allow at least a free trial, so if you find several that appear to meet your needs, it is wise to test them out before committing. 

    Platforms to consider using when collecting mobile data

    Resources

    Course

    Examples

    Guides

    Tools

    Jung, C. NOMAD, (2011). Mobile data collection systems a review of the current state of the field. Retrieved from website: https://www.alnap.org/help-library/mobile-data-collection-systems-a-review-of-the-current-state-of-the-field

    Kugler, K. (2012, January 24). The cellphone that keeps the water, and data, flowing. Retrieved from http://blog.nationalgeographic.org/2012/01/24/the-cellphone-that-keeps-the-water-and-data-flowing/ (archived link)

    Expand to view all resources related to 'Mobile data collection'

    'Mobile data collection' is referenced in: