51% of research projects are delivered as PowerPoint documents* *2009 Globalpark Market Research Software Survey from meaning ltd That's up from 48% in 2006 Just how important is PowerPoint to survey research? A question we asked: What percentage of projects currently involves the following deliverables or distribution methods to the client?
Online static reports
Digital dashboards Among larger research companies, it's already 57% What’s the problem with PowerPoint? The ubiquity of Powerpoint We learned… 2009 Globalpark Market Research Software Survey from meaning ltd If anything, I'm surprised it isn't more What are the viable alternatives? But what do we use PowerPoint for in market research? Criteria
easy to use
available now Prezi Instant Intelligence Reporting Dapresy Infotools Viewers Q InfoTrend for trackers InfoSwitch for loyalty studies Pizzazz in Research: Renewing the Rules of Engagement
Thursday 30th September 2010, Imperial College, London ASC One-day conference September 2010 Creating engaging results Loading PowerPoint… We also asked what improvements users wished for most in the analysis and reporting software they used PowerPoint again, but closely followed by web-browser analytical tools and dashboards Q These new tools all provide interactive visualisations with movement. Debriefs As the ‘deliverable’ As the report It's a presentation tool that is no longer being used to prepare presentations Presentations Presenting charts As a replacement for the report Presenting results Delivering bullet-point reports Distributing results Sending the presentation to people who don't attend the debrief Gapminder from Tim Macer Some viable antidotes to PowerPoint Instant Intelligence
Reporting 2. Build a crosstab: choose the variables 3. Pick a canned layout 4. Preview the cross-tab with the data you currently have 5. Now view it as a chart — but this chart doesn't do it for this question… 6. Try a different template until you are happy with the chart 7. Now get the software to save this as an editable Excel file 8. The charts are hooked to the data - through this is just snapshot of the current data while you are working to finesse your charts in Excel 10. Preview it to see what the end user will see it when published. The user now has drop-down menus to vary the variables and filters being viewed. 1. A dashboard microsite based on MR data — users can select reports, and do dashboardy things like choose different filters or variables 9 Next, you re-import the updated Excel workbook to use it as a template for creating all the reports with the variables and filters that you choose 12. IIR will publish to Excel, PowerPoint and even Xcelsius, as here. 11. Now you create the “reports” you want to view from the charts created. The folders and reports will form the tabs and dropdowns. an expert system for lay users to analyse survey data using advanced multivariate methods Rich metadata is used in the background to determine what kind of analysis is relevant as you select each question Q The ‘question’ here is set of related questions — Q understands how to display this as a grid of questions.
It will also use the metadata to determine what values or statistics are probably the most useful ones to display. Here it has automatically selected row percentages. Q will examine the metadata of the questions being presented to decide what is the best significance test to apply.
When testing batches of table, it uses the false discovery rate (that is, rather than testing at the 0.05 level of significance, it examines the data to find a more appropriate level of significance) Q's ‘smart tables’ feature lets you pick a question, then throw all the other variables at, and it will sort the resulting series of cross-tabs according to statistical significance The user just selects the variable, but behind the scenes, Q chooses the best significane tests to use, for example
With the multiple response data, Wilk’s Lambda is used to test for statistical significance
With a ranking crossed with a categorical variable, a Likelihood Ratio test will be conducted
With a crosstab of two categorical variables, Pearson’s Chi Square Test of Independence is conducted. There’s an easy tool for creating maps. Q looks at the metadata to determine what map to produce: here it is a correspondence map. The map also lets the user edit it, such as to take out ‘none of these‘ or cross it by itself. It redraws the map and if necessary will use a different method too Q understands time series and lets you re-calibrate your timelines — such as to make this monthly series into a weekly series As months shift to weeks, Q intelligently handles missing periods It also puts the series labels in a nice place! motion charts first developed by Gapminder.org ideal for animating time series data could be applied to MR trackers or any trend data now freely available as a part of Google chart tools as 'Google Motion Charts' This is Prezi A great new tool for creating presentations Web based - but can download an executable for Mac or PC Non-linear, non hierarchical Can incorporate graphics and Flash or MP4 movies very easily Apply order to chaos by defining a path Edward Tufte James Parsons Nick Southgate Tufte, E. PowerPoint is Evil, Wired Magazine, 2003
Tufte Edward, 2003. Essay: “The cognitive style of PowerPoint.” “Presentations largely stand or fall on the quality, relevance, and integrity of the content. If your numbers are boring, then you've got the wrong numbers. If your words or images are not on point, making them dance in color won't make them relevant. Audience boredom is usually a content failure, not a decoration failure.” Highly critical of PowerPoint’s weaknesses. Highlighed the fatal role a ‘PowerPoint culture’ played in allowing the Challenge Shuttle disaster to happen. Important knowledge was there, but was buried, filtered away and never reached senior management because PowerPoint had largely replaced the written report. Specific criticisms:
sequential nature: thought and ideas are not linear
Low information content - typically 40 words per slide = a few seconds silent reading
False hierarchies: often the most important information is at the lowest level in the smallest writing Parsons, J. 2004. “PowerPoint is not written in stone. Business communication and the lost art of storytelling.” Research Conference. MRS. London. PowerPoint is a useful medium but market researchers must be aware of its dangers It is a visual medium — landscape format, projected images—sits uneasily with facts and content
Ritualistic—a shared experience with certain expectations—but remove the deck and circulate in isolation, and the context is lost
Designed to be looked at, not read “So this is common practice in some organisations…
“Documents written in a medium designed to be looked at for immediacy rather than
read for detail, by convention vulnerable to ambiguities, invested with a mythical
authority, written using mechanisms of abridgement and abbreviation over which
there is no convention or consensus, are passed as key documents to persons not
present at the meeting of their presentation.” Southgate, N. 2006. “The cost of cliché charts: I never want to see again.” Research Conference, MRS, London. “The cost of cliché is an insipid one. Cliché fills spaces but empties minds. It makes what research does seem predictable and commonplace. It drives out real insight in favour of platitudes. It treats fertile areas of enquiry with glibness. It strips meaning from the provocative.” Critical of presentations and debriefs that rely heavily on cliché, on gratuitous illustration and on samey charts. Structural problems
low information content per slide
sequential nature makes comparisons difficult
difficult and time-consuming to produce
Problems of misuse
clichéd, boring and just not engaging
widely used not to support an oral presentation but as a report substitute
hard for users to search and navigate Conclusions It really doesn't have to be PowerPoint every time We need to stop abusing PowerPoint by using it as the principal deliverable: it is a presentation tool To make results engaging, users need to be able to engage with them Design matters The new tools can add the missing components of depth and movement Our responsibility as curators of content Moves data presentation from a two-dimensional view to a multi-dimensional one A true portal/dashboard tool for survey data that requries only conventional DP skills (and preferably, an eye for the visual too) With appropriate authority users can drill down to view individual verbatim comments — ideal for voice of the customer or mystery shopping surveys There is good support for time series data, with a built-in understanding of periods and smoothing, based on date alone The flexible and customisable charting is not based on MS Office. Drill-down actions can be defined that are approporiate to the content and what the user is likely to want to see in more detail Dapresy is strong on stakeholder reporting, where the view of the data varies according to the user's role and responsibilities Different data sources can easily be combined—case data and summarised data Compact, dashboard style charts like this, with several different kinds of data and presentation are notoriously difficult to create or update in PowerPoint or Excel, but are built-in to this software. A simple-to-use survey dashboard workflow which delivers dashboards and stakeholder reports in a variety of formats There are numerous problems Problems we face in research [Press ">" below to continue] Dear viewer,
Thanks for watching this Prezi. I hope you’ve found it useful. You are in control now. If you move your mouse pointer to the middle of the right edge of this window, you can zoom in and out. If you have a wheel on your mouse you can also zoom in and out. Click and drag to move around and revisit any part of this presentation.
If you've found this useful or have any feedback, please leave a comment.
London, UK http://www.meaning.uk.com Please visit our websiteSee the full transcript