New version SPSS will include ‘celebratory fireworks’ for significant results

An official press release has confirmed that the newest release of SPSS will be equipped with ‘performance-rewarding features’. The new installment of the popular data-analysis package will light up with song, dance and fireworks whenever a statistical test is significant. ‘We want to provide a package that is in line with the day-to-day experiences of researchers. We understand the pressure the publish, and the relief that is felt by many when those Stars of Significance appear in the results table. ’

The level of significance will determine the abundance of the celebrations. If the p-value is below 0.05, researchers will automatically hear what is described as ‘a cheerful tone’, according to a company spokesman. “But if your p-value is below 0.01, the software package will play a series of congratulatory videos, complimenting your experimental design and choice of analysis. And if it is very highly  significant, or below 0.001, your extra order of magnitude is rewarded by a lavish display of fireworks, clinking of champagne glasses and a showtune that plays ‘Tenure is here to stay’.

Dr. Hellst from the University of Ontario thinks it is a logical step: “Research is hard work. It can take months, sometimes even years, to collect the data. It’s such an anticlimax when are in your office, you run the analysis, the results are significant and the computer is completely and utterly silent. As if it doesn’t even care that my three-way ANOVA came out exactly the way I predicted. I’m so glad that the new edition of SPSS captures my feelings of elation in a suitable, yet professional, manner.”

Advertisements

5 comments

  1. Matus

    I guess you are all still oblivious to the cheat mode included since v17. Cheat mode can be activated by importing a .csv with a single cell containing value 42. Once activated, SPSS includes the option to run the data dredging wizard (Analysis -> Extras -> DD Wizard) or to generate your own dataset with pre-specified effects and significance (Analysis -> Extras -> Smeester Twister) .

    I bet your R can’t do that!

  2. Robyn

    This is both funny and truly sad– on multiple levels. In most cases, statistical significance testing is misapplied, inappropriate, and misleading–and the joyful visual fanfare will simply reinforce the worst tendencies of semi-literate and even experienced researchers and users.

    Two references to explain why statistical significance testing, including the reliance on p values, is so misguided.

    1. See Ronald Carver’s classic piece in the Harvard Educational Review “The Case Against Statistical Significance Testing” http://scholasticadministrator.typepad.com/thisweekineducation/files/the_case_against_statistical_significance_testing.pdf

    It’s an eye-opener.

    2. Read and enjoy the wonderful roundtable by a set of experience, veteran, highly-skilled statisticians discussing the field’s inherent flaws.

    “Standard Errors in Educational Assessment: A Policy Analysis Perspective ” Camilli Rutgers University Volume 4, No. 4 of Education Policy Analysis Archives http://epaa.asu.edu/epaa/v4n4.html

  3. Arthur Stabolidis

    I am also a little amused and disturbed. In 2012 I did an honours thesis on statistical reporting practices in psychology. It was actually fun and interesting! Gerd Gigerenzer and .Raymond Nickerson had the most interesting and productive things to say on the matter. John Hunter wrote a scathing piece in 1997 calling for a ban on NHST which was quite impassioned and amusing.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s