Fraud means ‘Science’. Or rather $cience and $cientism.

20 years ago John Ioannidis Published, "Why Most Published Research Findings Are False" in PLoS Medicine (2005). It is still relevant, even if it understates the issues by half. Ioannidis highlighted problems which have only blossomed and flourished beyond all control in the last 2 decades:

1. Small sample sizes: Many studies purposely use very small sample sizes, which distorts the data and provides false positives (the skew theme).

2. Small effect sizes: Small sample sizes will produce ‘small effect sizes’, which indicates that they are unreliable and statistically invalid (the stats fraud theme).

3. Large numbers of tested relationships: When many hypotheses are tested, the likelihood of finding a spurious ‘significant’ result increases (the shallow-analysis and misdirection theme).

4. Flexibility in designs, definitions, outcomes, and analytical modes: This allows researchers to ‘p-hack’ or selectively report findings that meet statistical significance (the ‘confirmation bias’ and significance fraud theme).

5. Financial and other interests and prejudices: Conflicts of interest obviously distort and bias research outcomes. How many reports have you read which state the author(s) biases, worldviews and funding (the corruption theme) ?

6. Publish or die: In competitive fields, there's pressure to publish narrative supporting ‘studies’ quickly, which leads to fraudulent or flawed research (the race-to-the-bottom-theme).

All of the above saturate ‘The Science’. The priority is of course money. When you follow the funding, you will find the ‘science’, much of it fiction.  More here