# Random sampling versus random allocation/randomization- implications for p-value interpretation

**URL:** <https://discourse.datamethods.org/t/random-sampling-versus-random-allocation-randomization-implications-for-p-value-interpretation/6177>\
**Category:** general\
**Created:** [November 6, 2022, 9:30pm UTC](https://discourse.datamethods.org/t/random-sampling-versus-random-allocation-randomization-implications-for-p-value-interpretation/6177 "2022-11-06T21:30:48Z")\
**Posts on this page:** 1\
**Showing post:** 55

<div class="post-metadata">

**Author:** ![davidcnorrismd](https://discourse.datamethods.org/user_avatar/discourse.datamethods.org/davidcnorrismd/32/3502_2.png) [@davidcnorrismd](https://discourse.datamethods.org/u/davidcnorrismd)\
**Post date:** [September 29, 2023, 9:35am UTC](https://discourse.datamethods.org/t/random-sampling-versus-random-allocation-randomization-implications-for-p-value-interpretation/6177/55 "2023-09-29T09:35:49Z")

</div>

> [@Pavlos\_Msaouel](#):
>
> In oncology, we have even reached the point of having high-profile “randomized non-comparative” trials [being published](https://pubmed.ncbi.nlm.nih.gov/37582240/). This is an oxymoron to methodologists,

I wonder if this oxymoronic phrase could help understand the culture of [Anomal Pharm](https://discourse.datamethods.org/t/anomal-pharm-randomized-good-statistical-bad/5086), where an oddly _non-statistical_ form of randomization seems to be advanced, in language such as:

> “And so the purpose of randomization is not necessarily to establish that one dose of the drug is superior to another dose in a _statistical_ way, or establish _statistical_ non-inferiority between the two doses, but really to have trials that are sufficiently sized so that we can interpret these dose-efficacy and dose-toxicity relationships, and use that information to guide overall decision-making.”  
> — Mirat Shah

Perhaps a similar spirit is revealed in a paper by Iasonos & O’Quigley [1], where we are instructed that

> The goal of randomization is to reduce biases due to known but uncontrollable factors such as prior treatment, age, gender and/or comorbidities.

 ![image](https://discourse.datamethods.org/uploads/default/original/2X/b/b33a0d1e249042dc87ac59c2789115c1ff3b7571.png)

1. Iasonos A, O’Quigley J. Randomised Phase 1 clinical trials in oncology. _Br J Cancer_. 2021;125(7):920-926. doi:[10.1038/s41416-021-01412-y](https://doi.org/10.1038/s41416-021-01412-y)

---

_[View the full topic](https://discourse.datamethods.org/t/random-sampling-versus-random-allocation-randomization-implications-for-p-value-interpretation/6177)._