Realizations in Biostatistics

Biostatistics, clinical trial design, critical thinking about drugs and healthcare, skepticism, the scientific process.

Thursday, March 18, 2010

Love p-values for what they are, don't try to make them what they're not : Applied Statistics

Love p-values for what they are, don't try to make them what they're not : Applied Statistics

Yet more perspective on the worshipped and/or abhorred p-value.
Posted by John Johnson at 10:07 AM
Newer Post Older Post Home
Subscribe to: Post Comments (Atom)

About Me

My Photo
John Johnson
I am a statistical consultant and Associate Director of Statistics at REGISTRAT-MAPI, specializing in late-phase research, observational trials, real-world outcomes, and weird problems.

I am interested in healthcare data, data mining, and open/distance education.

If you want to inquire about statistical consulting, please contact me at johndjohnson@gmail.com.

I work hard to be a good father to two children.
View my complete profile

Google+ Badge

Share

Tweet

Subscribe

Posts
Atom
Posts
Comments
Atom
Comments

Twitter Updates

Twitter Updates

    follow me on Twitter

    Popular Posts

    • Followon to polling post
      Seems like the guy at fivethirtyeight.com is pretty on top of polling methods, including modeling of the polling data (weighting by past pe...
    • O'Brien-Fleming designs in practice
      It seems that the O'Brien-Fleming design is the most popular of all group sequential clinical trial designs. This particular strategy re...
    • Matrix square roots in R
      There are a couple of ways to do matrix square roots in R. The first is through eigendecomposition: # a is the matrix for which we want ...
    • Analysis of the statistics blogosphere
      My analysis of the statistics blogosphere  for the Coursera Social Networking Analysis class is up. The Python code and the data are up at ...
    • My O'Brien-Fleming design is not the same as your O'Brien-Fleming design
      I know this discussion is a little technical, and nonstatisticians can probably skip this, but I hope that a statistician struggling with th...

    Recent drug development and statistics content

    • Eye on FDA
      Weekly Roundup 5-17-13 - It has been a busy May. From working in the yard to transform it from looking like property on which Boo Radley might have a house to putting on panel dis...
      17 hours ago
    • FDA newsfeed
      FDA seeks preliminary injunction against New York fish manufacturer - The U.S. Department of Justice, on behalf of the U.S. Food and Drug Administration (FDA), is seeking a preliminary injunction in the U.S. District Court fo...
      17 hours ago
    • In the Pipeline
      A Little Ranbaxy Example - Compare and contrast. Here we have Krishnan Ramalingam, from Ranbaxy's Corporate Communications department, in 2006: * Being a global pharmaceutical maj...
      23 hours ago
    • FDA Law Blog
      Third Circuit Holds that the FDC Act Preempts Class Action Regarding Absence of Trans Fat and Cholesterol Reducing Effect - By RiĆ«tte van Laack – On May 9, 2013, the U.S. Court of Appeals for the Third Circuit affirmed a District Court’s decision that a state law class action conc...
      1 day ago
    • Junk Charts
      More power brings more responsibility - Nick C. on Twitter sent us to the following chart of salaries in Major League Soccer. (link) This chart is hosted at Tableau, which is one of the modern visu...
      2 days ago
    • WSJ.com: Health Blog
      How Well Does Your Doctor Communicate with You? - Patients' complaints about doctors are finally being taken seriously, as poor communication is increasingly understood to be at the root of many of health ...
      1 month ago
    • Pharmalot
      Should FDA Have Approved Lilly Alzheimer’s Agent? - Late last week, the FDA approved an imaging agent from Eli Lilly for detecting beta-amyloid plaques in patients with cognitive impairment and who are being...
      1 year ago
    • Statistical Modeling, Causal Inference, and Social Science
      Visualizing election polls - A colleague points me to these supremely ugly pie-like graphs by Richard Riesenfeld and Geoff Draper. On the other hand, who am I to say they're ugly? I'...
      4 years ago

    Personal Links

    • My family blog
    • My other opinions

    Statistics and Data Analysis

    • Stats blogs
    • R-bloggers
    • The Do Loop
    • The SAS dummy
    • Revolution Analytics
    • Effective Graphs
    • Statistics Forum
    • Chance News at Dartmouth
    • Epidemiology Methods Blog
    • Flowing Data
    • Gapminder
    • Introduction to Bayesian Thinking
    • John Cook
    • Junk Charts
    • Lies and Stats
    • Mathematical Epidemiology
    • Radford Neal
    • Statistical Modeling, Causal Inference, and Social Science
    • UN Statistics Division Common Database

    R

    • Revolutions
      Because it's Friday: Compendium of Logical Fallacies
      14 hours ago
    • R bloggers
      R 3.0.1 released
      16 hours ago
    • One R Tip A Day
      R 3.0.0 is released!
      1 month ago
    • Cerebral Mastication
      Installing & Debugging ODBC on Mac OS X
      3 months ago
    • SAS and R
      Example 10.8: The upper 95% CI is 3.69
      5 months ago
    • R-Chart
      Github Follower Graph with R
      1 year ago
    • Programming R - Articles
      Animations in R
      1 year ago
    • Learning R
      Consultants’ Chart in ggplot2
      2 years ago

    Drug development

    • Pharma IQ
    • FDA home page
    • FDA Critical Path
    • Clinical Trials Today
    • PhRMA
    • Ask Cato
    • Cato Research

    Clinical trials

    • On biostatistics and clinical trials
    • Good Clinical Trial Simulation Practices
    • PhRMA's clinical study results
    • Government's registry of clinical trials

    Labels

    adaptive trials (11) alternative medicine (1) alzheimer's disease (2) analysis (1) ANOVA (1) antibiotics (1) Ask Cato (1) autism (1) Avandia (2) Bayesian statistics (13) bias (1) big data (2) biomarkers (1) biostatistics (11) blinding (4) blogging (9) Blogkeeping (2) books (2) careers (3) causal inference (3) central nervous system (1) CETP inhibitors (1) chaos (1) classes (3) clinical trials (36) comparative effectiveness (2) compliance (1) consulting (3) contests (1) correlation (1) coursera (2) crf (1) critical path initiative (1) critical thinking (4) data cleaning (1) data collection (3) data mining (5) databases (1) delta method (1) drug development (2) drugs (2) echinacaea (1) Ed Silverman (1) education (5) eli lilly (1) email (1) EMEA (1) ENHANCE (2) epidemiology (2) FDA (6) free (1) Google (1) graduate school (1) graphics (6) group sequential (4) Hal Varian (1) humor (1) Inference for R (2) intention to treat (1) interpretation (4) John Tukey (1) joint statistical meetings (1) JSM (2) JUPITER (1) Lan-DeMets (1) leadership (4) learning (2) learning from data (1) links (1) loess (2) lying with statistics (9) machine learning (3) Mantel-Haenszel (1) marketing (1) mathematics (2) matrix (2) MedDRA (1) metaanalysis (1) Microsoft Excel (1) Microsoft Word (2) missing data (1) modeling (2) modern regression (1) mooc (4) multiple comparisons (1) multiple sclerosis (1) natural language processing (2) networking (1) neurology (1) NNT (3) noninferiority (2) o'brien-fleming (1) observational (2) odds (1) Other blogs (7) p-values (2) PD (1) Peter Rost (2) pharmaceutical industry (3) pharmacogenomics (2) Pharmalot (2) Phase 2 (1) phase 3 (3) Phase I (1) PK (1) politics (6) polls (4) postmarketing (1) practice of statistics (1) Presentations (2) privacy (1) probability (1) professionalism (3) proof (1) publishing (1) Python (2) R (20) random number generation (1) randomization (3) relationships (2) risk (1) RStudio (1) RTF (2) safety (7) SAS (8) SAS programmer (1) science (1) simple statistics (1) simulation (1) skepticism (3) smoothing (1) social network analysis (4) statins (1) statistical analysis plan (1) statistical leadership (1) statistical programming (8) statistical significance (3) statistics (28) statistics forum (2) statistics without borders (1) surveys (1) t test (1) tables (1) Taylor series (1) TGN1412 (2) torcetrapib (1) twitter (1) Type I error (2) Ultraedit (2) vaccination (2) visualization (2) waste (4) writing (1)

    Blog Archive

    • ►  2013 (10)
      • ►  April (2)
      • ►  March (4)
      • ►  February (4)
    • ►  2012 (28)
      • ►  December (1)
      • ►  November (5)
      • ►  October (4)
      • ►  September (2)
      • ►  August (8)
      • ►  May (2)
      • ►  April (2)
      • ►  March (2)
      • ►  January (2)
    • ►  2011 (26)
      • ►  December (7)
      • ►  October (1)
      • ►  September (3)
      • ►  August (2)
      • ►  June (2)
      • ►  May (1)
      • ►  March (1)
      • ►  February (1)
      • ►  January (8)
    • ▼  2010 (50)
      • ►  December (3)
      • ►  November (2)
      • ►  October (3)
      • ►  September (8)
      • ►  August (8)
      • ►  July (5)
      • ►  June (5)
      • ►  May (4)
      • ►  April (4)
      • ▼  March (5)
        • Love p-values for what they are, don't try to make...
        • Odds are, you need to read this
        • Observation about recent regulatory backlash again...
        • When a t-test hides what a sign test exposes
        • Another strike against p-values
      • ►  February (3)
    • ►  2009 (28)
      • ►  December (1)
      • ►  November (2)
      • ►  October (1)
      • ►  September (1)
      • ►  August (1)
      • ►  July (3)
      • ►  June (5)
      • ►  May (6)
      • ►  April (2)
      • ►  March (3)
      • ►  January (3)
    • ►  2008 (47)
      • ►  December (2)
      • ►  November (4)
      • ►  October (5)
      • ►  September (7)
      • ►  August (5)
      • ►  July (4)
      • ►  June (3)
      • ►  May (7)
      • ►  April (2)
      • ►  February (2)
      • ►  January (6)
    • ►  2007 (32)
      • ►  December (1)
      • ►  November (3)
      • ►  September (3)
      • ►  August (9)
      • ►  July (2)
      • ►  June (2)
      • ►  May (2)
      • ►  April (3)
      • ►  March (2)
      • ►  February (3)
      • ►  January (2)
    • ►  2006 (3)
      • ►  December (3)

    Site Meter

    Powered By Blogger
    Awesome Inc. template. Powered by Blogger.