Skip to main content
SAGE
Search form
  • 00:00

    [MUSIC PLAYING][A BRIEF GUIDE TO RANDOMIZED CONTROLLED TRIALS]

  • 00:07

    ANOUK RIGTERINK: My name is Anouk Rigterink.[Anouk Rigterink, PhD] I'm a postdoc at the Universityof Oxford, and I'll be talking about randomized controlledtrials.[HOW DID YOU BECOME INTERESTED IN USING THIS METHOD?]I became interested in it, becauserandomized controlled trials are really the gold standard

  • 00:30

    ANOUK RIGTERINK [continued]: in certain types of research.If you want to talk about whether something--say a policy or an intervention--causes an outcome, rather than is just correlated or relatedto it, then randomized controlled trialsare the first go-to method to establish this.[WHAT KINDS OF RESEARCH QUESTIONS WOULD THIS METHOD BE

  • 00:52

    ANOUK RIGTERINK [continued]: SUITABLE FOR?]So the kind of research questionsthat you can answer with a randomized controlled trial?Actually, there's one very generic research question.Does this intervention achieve this outcome?So for instance, one research questionI'm interested in-- does monitoring of the forest

  • 01:15

    ANOUK RIGTERINK [continued]: by a community decrease deforestation?But we can also think about, does giving people job traininghelp them get a job?Does making women and girls aspireto more lead them to start up more and more successfulbusinesses?Literally, any policy that can be

  • 01:38

    ANOUK RIGTERINK [continued]: given to some individuals, groups, villages, but notothers, is a potential thing that youcan investigate using a randomized controlled trial.[WHAT KINDS OF DATA CAN BE GATHERED OR ANALYZED WHEN USINGTHIS METHOD?]Usually, when you do a randomized controlled trial,

  • 02:02

    ANOUK RIGTERINK [continued]: you collect your own data, often through a survey of some kind,so a questionnaire that you put to peoplewho have either received the treatment,or not receive the treatment.Before you can start analyzing the datain randomized controlled trials, there is a lot of planning

  • 02:23

    ANOUK RIGTERINK [continued]: that goes on beforehand.Because in a randomized controlled trial,it is truly an experiment in the sense that the researchdetermines who gets a certain treatment and who doesn't.So there is a lot of planning going onin a randomized controlled trial.First, you have to find out what the situation waslike before you started.

  • 02:44

    ANOUK RIGTERINK [continued]: So that's called a baseline.So you will go, and you will do your survey.And you will find out from your individuals,your potential recipients of the treatment,villages, et cetera, what things were like before you give themthe treatment.Then you're going to, as a researcher,give or arrange to give particular individuals

  • 03:09

    ANOUK RIGTERINK [continued]: the treatment, and other individuals not.So for a forestry treatment, our units are in villages.And in some villages, people are trained to monitor the forest,but in other villages they're not.And the way we decide who gets the treatmentand who doesn't is random.That's why it's a randomized controlled trial.

  • 03:30

    ANOUK RIGTERINK [continued]: We didn't make any decisions, we hada computer on the basis of a random number generatordecide which villages got the treatment and which are not.So we don't do this because we liketo leave all responsibility to the computer,we do this because, by being random,we can make sure we can make an expectation we can be sure

  • 03:51

    ANOUK RIGTERINK [continued]: that the treatment we give isn't related to any other factors,in this case the community, that may already be there.So we are not consciously picking communitiesbecause we think, oh, these communities monitoring willbe really successful.We're not picking communities because wethink, oh, deforestation is really bad in this village

  • 04:12

    ANOUK RIGTERINK [continued]: already.Because if we did that, then we wouldn't besure what causes our outcome.Is there less deforestation because communitieswere just better to begin with?Is there more deforestation because communitieswere just worse to begin with?It is random.So in expectation, the villages that get the treatmentand the villages that don't get the treatment, in expectation,

  • 04:34

    ANOUK RIGTERINK [continued]: they are the same.[CAN YOU GIVE AN EXAMPLE OF WHERE YOU HAVE USED THIS METHODAND WHAT YOU LEARNED IN THE PROCESS?]What I've learned from using this method,personally, is that when you read the report of a randomizedcontrolled trial or an academic article that

  • 04:56

    ANOUK RIGTERINK [continued]: reports on a randomized controlled trial,it all reads quite slick.We planned the treatment.It was executed.We did our baseline, we find out what was happening before.We did our end line, we found outwhat happened after the treatment.In practice though, because this is an experimentand the researcher has to do a lot of work to make sure

  • 05:19

    ANOUK RIGTERINK [continued]: the treatment gets implemented, so in reality, thereis a lot of messiness going on.There's a lot of work going into this.So one thing, for instance, very simple,we wanted to pick certain villages.And they had to be adjacent to the forest,and they had to not border one another.

  • 05:41

    ANOUK RIGTERINK [continued]: Now, this sounds simple enough, right?But we did this in Uganda, so we hadto get a list of all Ugandan villages what we got.But then, of course, not all villages are on there.Some have been split.Some have disappeared already.Every name can be spelled just about a milliondifferent ways--so just all these little practical problems.I think my main learning point is that, in practice, RCTs

  • 06:07

    ANOUK RIGTERINK [continued]: are a lot messier and a lot harderthan it seems when it's written up neatly.[WHAT ADVICE DO YOU HAVE FOR STUDENTS LOOKING TO USE THISMETHOD?]I think, for a student, it would bequite difficult to do a randomized controlled trialyourself, because you have to collect your own data,

  • 06:31

    ANOUK RIGTERINK [continued]: you have to make sure that there is a treatment being done.But of course, because the researcher'sresponsible for implementing that treatment,it also gives the researcher an ethical obligation.So you have to get ethical approval.In practice, these things cost a lot of money.They cost a lot of time.You have to make a lot of partnerships, potentially.

  • 06:52

    ANOUK RIGTERINK [continued]: So I don't think, as a student--maybe short of someone who is doing a PhD--as a student, it would be really difficult to doa randomized controlled trial by yourself.So that is the bad news.On the one hand, the good news isthat there are a lot of people doingrandomized controlled trials.And as I mentioned before, there's a lot of work involved.So chances are, someone is doing a randomized controlled trial.

  • 07:13

    ANOUK RIGTERINK [continued]: They are looking for assistants.So if you want to get experience with doinga randomized controlled trial, tryto find someone who's doing it.Offer your assistance.Maybe you have an internship, something like that.Get some experience.And then maybe if you choose to go on and doa PhD or work in a policy environment,you can do one of your own.[MUSIC PLAYING]

Video Info

Publisher: SAGE Publications Ltd

Publication Year: 2019

Video Type:Tutorial

Methods: Randomized controlled trials

Keywords: deforestation; internships; intervention studies; policy analysis; villages

Segment Info

Segment Num.: 1

Persons Discussed:

Events Discussed:

Keywords:

Abstract

Anouk Rigterink discusses randomized controlled trials and how to utilize this method to gather and analyze data.

Looks like you do not have access to this content.

A Brief Guide to Randomized Controlled Trials

Anouk Rigterink discusses randomized controlled trials and how to utilize this method to gather and analyze data.

Copy and paste the following HTML into your website