A A A
Manual Event (Session ID,DateTime,Event) = mqh5mudc5opmn75rik9h43nr15 20210918033736 /sc/communities/research-design-2/

Research design

Share:
Whtasapp
主題討論

Hello! I’m Dr. MEL. Nice to meet you! I’m absolutely excited that we are going to start this beautiful learning journey together and I’m sure you will find it fruitful and fulfilling. Feel free to share your thoughts here!

最近的活動討論

Got questions? Simply post them and let’s discuss! 

討論:General
Toby.wong
2019年12月30日
Research design

I would like to know more about research design. Any suggestions?

Reply
Like
Share
Whtasapp
Michael Lo
2019年12月30日

Baseline A Treatment B Baseline A
O-O-O-O-O-X-X-X-X-X-O-O-O-O-O-O

回覆
Like
分享
Whtasapp
Gladys.Kwong
2019年12月30日

Nice. Can you also elaborate in notation?

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

This design needs multiple observations of a single individual. The target behavior of a single individual is established over time and is referred to as a baseline behavior. The baseline behavior is assessed, the treatment provided and then the treatment is withdrawn.

回覆
Like
分享
Whtasapp
Gladys.Kwong
2019年12月30日

Great.

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

I am happy to share with you! The last but not least is Single-Subject Designs. Let take A-B-A Single-Subject Design as an example.

回覆
Like
分享
Whtasapp
Gladys.Kwong
2019年12月30日

Thanks a lot.

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

Sure. This involves random assignment of participants to 4 groups. It’s a special case of a 2X2 factorial design. Pre-tests and treatments are varied for the four groups. All groups receive a post-test.
Group A R------O------X------O
Group B R------O--------------O
Group C R --------------X-----O
Groupd D R---------------------O

回覆
Like
分享
Whtasapp
Gladys.Kwong
2019年12月30日

I see. Can you elaborate more about it?

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

No. We still have Solomon Four-Group Design.

回覆
Like
分享
Whtasapp
Gladys.Kwong
2019年12月30日

That’s all for different kinds of True Experimental Designs?

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

Yes you are right.
Group A R------X------O
Group B R--------------O

This is the case.

回覆
Like
分享
Whtasapp
Gladys.Kwong
2019年12月30日

I heard about it and it’s popular. This design controls for any confounding effects of a pre-test. The participants are assigned to groups randomly. The treatment is only given to experimental group. We measure both groups in post-test.

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

You are welcome! The second one is Post-Test-Only Control-Group Design.

回覆
Like
分享
Whtasapp
Sunny Lee
2019年12月30日

Thank you very much for your explanation.

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

Yes absolutely!

回覆
Like
分享
Whtasapp
Sunny Lee
2019年12月30日

Let me try to illustrate by notations. Is it…
Group A R------O------X------O
Group B R------O---------------O

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

This design needs random assignment of participants to 2 groups. Both groups are administered both a pre-test and post-test but the treatment is for experimental Group A.

回覆
Like
分享
Whtasapp
Sunny Lee
2019年12月30日

What does it mean?

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

The first one I want to talk about is Pre-Test-Post-Test Control-Group Design.

回覆
Like
分享
Whtasapp
Sunny Lee
2019年12月30日

Thanks for your reply!

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

Yes, apart from pre-experimental and Quasi-experimental designs, there are True Experimental and Single-Subject Designs.

回覆
Like
分享
Whtasapp
Sunny Lee
2019年12月30日

Are there any other types of research designs?

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

Sure. Please see below
Group A O--O--O—O--X--O--O--O—O
------------------------------------------------
Group B O--O--O--O--O--O--O--O--O

回覆
Like
分享
Whtasapp
Iris Man
2019年12月30日

Can you express more clearly by using notations?

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

Its modification in which two groups of participants, not randomly assigned, are observed over time. A treatment is administered to only one of the groups (i.e., Group A).

回覆
Like
分享
Whtasapp
Iris Man
2019年12月30日

What’s it?

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

Yes, the last but not least is Control-Group Interrupted Time-Series Design.

回覆
Like
分享
Whtasapp
Iris Man
2019年12月30日

I see. Thank you so much! Any others?

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

Yes. It can be expressed as:
Group A O—O—O—O—X—O—O—O—O

回覆
Like
分享
Whtasapp
Iris Man
2019年12月30日

I remember that we record measures for a single group both before and after a treatment.

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

Yes it is.

回覆
Like
分享
Whtasapp
Iris Man
2019年12月30日

Is Single-Group Interrupted Time-Series Design also a kind of Quasi-Experimental Design?

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

Yes. You are right.

回覆
Like
分享
Whtasapp
EmilyChiu121
2019年12月30日

Yes…Is it
Group A O------X------O
-----------------------------
Group B O---------------O

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

It’s a popular approach to quasi-experiments. Experimental group (A) and control group (B) are selected without random assignment. Both of them take a pre- and post-test and only the experimental group receives the treatment.

回覆
Like
分享
Whtasapp
EmilyChiu121
2019年12月30日

Yes! I used it before.

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

First, Nonequivalent (Pre-Test and Post-Test) Control-Group Design.

回覆
Like
分享
Whtasapp
EmilyChiu121
2019年12月30日

I learnt several years ago but I’m not quite familiar with them.

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

Shall we talk about Quasi-Experimental Designs?

回覆
Like
分享
Whtasapp
EmilyChiu121
2019年12月30日

Yes please.

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

Yes actually I just mentioned some of them. If you are interested, I can write more here.

回覆
Like
分享
Whtasapp
EmilyChiu121
2019年12月30日

The above designs are ‘pre-experimental designs’. Are there any kinds of research designs?

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

My pleasure.

回覆
Like
分享
Whtasapp
Toby.wong
2019年12月30日

Great. Thank you very much for giving lots of examples!

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

The exception is that the nonequivalent comparison group received a different treatment.
Group A X1------------O
----------------------------
Group B X2------------O

回覆
Like
分享
Whtasapp
Toby.wong
2019年12月30日

Could you explain more?

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

We use the same procedure as the Static Group Comparison I have just mentioned as above, with an exception.

回覆
Like
分享
Whtasapp
Toby.wong
2019年12月30日

Oh, what is it?

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

Last but not least, Alternative Treatment Post-Test-Only With Nonequivalent Groups Design.

回覆
Like
分享
Whtasapp
Toby.wong
2019年12月30日

Thank you very much…

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

It is used after implementing a treatment. Researchers choose a comparison group and provides a post-test to both the experimental and comparison group(s) after the treatment.
Group A X ------------O
----------------------------
Group B ---------------O

回覆
Like
分享
Whtasapp
Toby.wong
2019年12月30日

What is it for?

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

Next, Static Group Comparison or Post-Test-Only With Nonequivalent Groups.

回覆
Like
分享
Whtasapp
Toby.wong
2019年12月30日

Nice. Any others?

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

The next one is One-Group Pre-Test-Post-Test Design which includes a pre-test measure followed by a treatment and a post-test for a single group.
Group A 01 ------X------02

回覆
Like
分享
Whtasapp
Toby.wong
2019年12月30日

Thank you very much!

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

Yes of course!

回覆
Like
分享
Whtasapp
Toby.wong
2019年12月30日

May I know more about the other types?

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

First, One-Shot Case Study. This design involves an exposure of a group to a treatment followed by a measure.
Group A X ------------O

回覆
Like
分享
Whtasapp
Toby.wong
2019年12月30日

Sure.

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

You are welcome. Now I am going to explain research designs in detail.

回覆
Like
分享
Whtasapp
Toby.wong
2019年12月30日

I see. Thank you very much for your illustration in advance.

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

You will see that there is a separation of parallel rows by a horizontal line. It indicates that comparison groups are not equal or equated by random assignment. No horizontal line between the groups displays random assignment of individuals to treatment groups.

回覆
Like
分享
Whtasapp
Toby.wong
2019年12月30日

Okay 

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

And random assignment is indicated with a symbol R.

回覆
Like
分享
Whtasapp
Toby.wong
2019年12月30日

I see…

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

The left-to-right dimension refers to the temporal order of procedures in the experiment. It is indicated with an arrow sometimes.

回覆
Like
分享
Whtasapp
Toby.wong
2019年12月30日

Yes. Yes…

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

X’s and O’s in a given row are applied to the same specific persons. X’s and O’s in the same column or placed vertically relative to each other, are simultaneous.

回覆
Like
分享
Whtasapp
Toby.wong
2019年12月30日

Ok…

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

O is what we can observe or measure on an instrument.

回覆
Like
分享
Whtasapp
Toby.wong
2019年12月30日

Yes…

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

X indicates an exposure of a group to an experimental variable or event and we are going to measure the effects.

回覆
Like
分享
Whtasapp
Toby.wong
2019年12月30日

Sure, thanks a lot.

回覆
Like
分享
Whtasapp
Michael Lo
2019年12月30日

Yes, but before I list some example, let me use the notations to illustrate the designs.

回覆
Like
分享
Whtasapp
最新討論帖
更多的討論帖
Hi MEL, I wonder is there any suicide prevention service in Hong Kong?
Hi MEL, I wonder is there any suicide prevention service in Hong Kong?
strength L
2021年8月28日
3個回覆,1個月前
Like
分享
Whtasapp
Following the previous question, how can I prevent such odd data from occurring in the future?
Following the previous question, how can I prevent such odd data from occurring in the future?
Snowy Cheng
2021年8月28日
3個回覆,1個月前
Like
分享
Whtasapp
Dear MEL professionals, would you please help me out with the odd data? For example, in a survey about low-income parents, the column of age has some strange values, such as 120.
Dear MEL professionals, would you please help me out with the odd data? For example, in a survey about low-income parents, the column of age has some strange values, such as 120.
Snowy Cheng
2021年8月27日
3個回覆,1個月前
Like
分享
Whtasapp
Where can I find the list of SSCI journals in the field of child welfare?
Where can I find the list of SSCI journals in the field of child welfare?
metacarpal
2021年8月27日
3個回覆,1個月前
Like
分享
Whtasapp
精選討論帖
Hi, I realized that it seems the concept “outcome” is different from the term “output”. Taking the grandparent training project as an example, what is the “outcome’ referable?
Hi, I realized that it seems the concept "outcome" is different from the term "output". Taking the grandparent training project as an example, what is the “outcome’ referable?
CouldYip
2021年8月26日
3個回覆,1個月前
Like
分享
Whtasapp
I have designed a large-scale survey about the life satisfaction of low-income families in Hong Kong. Is it appropriate to use the electronic survey through mobile phones?
I have designed a large-scale survey about the life satisfaction of low-income families in Hong Kong. Is it appropriate to use the electronic survey through mobile phones?
Realistic
2021年8月20日
3個回覆,1個月前
Like
分享
Whtasapp
Questions about program evaluation and impact assessment (PE & IA)?
Dear MEL participants, Hello everyone! We are going to meet Prof. Aron Shlonsky this week (17&18 Oct). Do you have any problems or issues in program evaluation and impact assessment? Pleas ...
meladmin
2019年10月15日
0個回覆,23個月前
Like
分享
Whtasapp
Question about the programme evaluation
Can someone recommend me a basic guide book for programme evaluation? We are an NGO in mainland China and providing services for young drug abusers in rural China. We learnt about your pro ...
Jessica
2019年10月9日
4個回覆,23個月前
Like
分享
Whtasapp