Questions tagged [research-design]
Methods and tips for designing and conducting effective research.
135
questions
3
votes
3
answers
278
views
What are several examples of survey questions which have response options designed as mutually exclusive when they are not?
Statisticians and Data Analysts sometimes draw conclusions from surveys given to people.
Often times, there is a question, and there are 2 to 5 options to choose from if you are a survey respondent.
...
3
votes
0
answers
39
views
+250
The probability of making a Type S error, and the average amount of magnification (type M error) as a function of power
When you filter on only viewing results with p<alpha, some of those "statistically significant" results go in the wrong direction. This is called a Type S error ("S" for sign). ...
- 18.3k
0
votes
0
answers
15
views
missing data without baseline characteristics measurements in one survey data
Suppose I obtained two survey data $D_1,D_2$. (I did not design surveys and I am told to conduct analysis. The survey sample is not a probability sample but a "census" of convenient sampling....
- 1,335
2
votes
1
answer
37
views
Post-stratification with missing subpopulations in the survey
I have some survey data on a population described by age, gender, and weight. It’s quite skewed so I want to reweight it to a known target population (a larger ...
- 1,002
1
vote
0
answers
33
views
How to compare AI model with other models for research purposes
If I have X and Y datasets, I will apply meta-learning few-shot-learning by training the model on the X dataset and applying meta-testing using the Y dataset. While I compare my result with other ...
- 33
0
votes
1
answer
15
views
Assessing the Validity of Conclusions Based on Limited Findings
Introduction: Archaeological research often involves drawing conclusions about past societies and cultures based on a limited number of artifacts or findings. In some cases, such as the study of ...
- 101
1
vote
1
answer
20
views
Right way of comparing my model's result with state of the arts
I know that I have to use the same datasets and experimental settings. In this way, I don't have to run their codes. What if I used different datasets? Is it OK to use their code from GitHub to get ...
- 33
0
votes
0
answers
16
views
How to group participants with nested variables (divisive clustering analysis)
Thanks in advance for the help!
Our goal is to generate groups of participants based on their vowel systems. Their vowel systems contain two nested variables:
The vowel class themselves (of which ...
- 1
0
votes
0
answers
13
views
If hypotheses get rejected, should one recraft the theoretical framework to better highlight the results or just report the results?
I developed a theoretical framework, based on signalling theory and further developed testable hypothesis. However, my hypotheses are rejected, results suggest the opposite. Countersignalling theory ...
- 205
0
votes
0
answers
9
views
Experiment terminology: control participants invertedly receive treatment
I have spent a solid hour to remember with the technical term for this. Suppose you have a between-subjects lab design with one control and two treatment conditions. What is the term to describe when ...
- 111
0
votes
0
answers
9
views
Difference-in-difference with border areas: how to create indicator for matched area?
I'm trying a DiD by comparing adjacent districts which lie on two sides of state border, before and after a policy. The scheme is following: state A introduced a policy in year t, while state B (...
- 1
0
votes
0
answers
20
views
obtaining the unit of observation for critical review of a research design
Hi I am writing a critical review for one of my assignments, and I am struggling to extract the unit of observation.
I have written a critical review for one of my assessments at university which is a ...
- 11
1
vote
0
answers
42
views
Analysis with data from different sources
I have data from 3 different sources, measuring different variables for different samples taken from the same population (a country). All of the data is from country-wide studies and should be ...
- 93
0
votes
0
answers
13
views
Can we do some causal conclusions if the mediator follows the dependent variable?
We have a survey consisting of two waves ($N=4000$).
In Wave 1, we ask block of questions $M_1$.
In Wave 2, we ask blocks of questions $W$, $Y$, $M_2$
The order of questions is as follows:
$W \...
0
votes
0
answers
10
views
How can I create a similarity network from a point cloud in vector space for classification purposes?
Are there any relevant techniques or approaches, such as k-nn graph or feature combination, that could be used to establish a similarity distance to create a graph to be used for gnn node ...
- 333
0
votes
0
answers
20
views
Selecting correct study design and statistical analysis for my project
I am analyzing risk factors associated with surgical site infections where we want to have a 3:1 case-control study. Other than some brief lectures in my classes, the only thing we were taught about ...
- 1
1
vote
0
answers
7
views
Disproportionate stratified sampling that follows proportional trend?
I'm conducting a study across four counties to determine which of four different methods is most effective in detecting nutria. 60 sites will be randomly selected within a 5-meter buffer of any open ...
- 11
1
vote
0
answers
74
views
How to analyze experimental data - 3x2 factorial design?
I am working on an experiment but am confused about what analyses I should undertake (I've picked the project up part way through). I’ve not been involved in experimental work in the past and so am ...
- 11
1
vote
0
answers
23
views
Can paired t-test be used when data comes from different task conditions at a single time point?
In my experiment a single group of participants performed a cognitive task with two interleaved types of blocks (A and B). Blocks A and B differed in terms of the type of stimuli used but were ...
- 11
2
votes
1
answer
50
views
Is multivariate ordinal regression suitable for my reserach?
I want to test what leads to the support for two policies (Likert scale ß disagree to strongly agree). There are independent variables, some of which are demographic and two are of interest (...
- 21
0
votes
0
answers
30
views
Example studies that use inferences from cross-validation of the entire dataset
I am performing a study where I perform inferences based on cross-validation metrics that use the entire dataset. My reasoning behind this is (a) my dataset is small and imbalanced and (b) this is ...
- 351
2
votes
1
answer
34
views
Is a mixed-model really required?
I've got data from a retrospective cohort analysis. There are two groups that differ by a certain baseline characteristic, who were all exposed to an event (surgical procedures). The outcome of ...
0
votes
0
answers
11
views
Building a research model/path analysis
I'm conducting psycho-oncology research. I am struggling with my model. Could you please help me if the model could look like this?
Moreover, I'd like to check the model with the path analysis in AMOS ...
0
votes
0
answers
27
views
Wilcoxon Signed-Rank Test N>300
There were similar questions here but as a person with little knowledge about statistics I couldn't understand any of it. I was looking for critical value tables that reached up to to higher amounts ...
0
votes
1
answer
46
views
Questions Regarding Sampling Bias
I'm taking a course in R: "Data Analysis in R" on Coursera, and I came across this question during the lecture:
A retail store considering updates to their credit card policies randomly ...
1
vote
0
answers
25
views
Advice on study design involving longitudinal clusters and measurements
I'm conducting an analysis on several hundred thousand individuals to understand if there are longitudinal differences in an outcome across clusters of the individuals. My outcome is healthcare cost (...
- 691
0
votes
0
answers
35
views
setting hypothesis based on 2 independent variable
Normally, we have one main IV and other covariates and show relationship with DV. Based on the main IV we conceptualize the research question and hypothesise. Say, Income (IV), Mental Health (DV) plus ...
- 1
1
vote
1
answer
177
views
Pooled cross-section or very highly unbalanced panel data?
Suppose I observe high school graduates applying to a certain college.
I want to model the probability to get admitted on the basis of individual characteristics (cognitive ability, test score, GPA, ...
- 1,928
0
votes
2
answers
78
views
What models can be used to prove causality?
Right now I am looking into analyzing the factors of crime in an economics paper. I currently have a multiple OLS regression set up that is pretty standard:
Y = B0 + B1(Variable) + B2(Variable) + B3(...
0
votes
0
answers
16
views
Estimate effect of intervention with only partial comparison group
I work in public policy, and our goal is to evaluate the effectiveness of an intervention that prevents a type of violence. The intervention has been running for a couple of years, and of those who ...
- 21
0
votes
0
answers
13
views
Is Exhaustive Theoretical Understanding Neccesary to do Research on Transformers?
I'm trying to come up with an original research topic for my thesis that tackles an application or open problem in Transformer based deep learning. After taking time to get an overview of the ...
- 113
5
votes
2
answers
252
views
Do I need to test for autocorrelation or normality assumption if I am running the regression with standard errors?
I used OLS regression to estimate a relationship between X and Y with a couple of control variables. However, when I tested for heteroskedasticity with Breusch–Pagan/Cook–Weisberg test, my residuals ...
- 205
3
votes
3
answers
301
views
What does the phrase "adjusted by" mean?
"Matching of one group with specific exposures or outcome with a
comparable group is frequently used in medical research to investigate
the association between exposure and outcome." Ref
...
- 1,801
0
votes
0
answers
10
views
Design for retrospective standardized testing and employment outcome study
we (student group) have been given a project involving about 10 years of corporate employment data, including variables on events like hiring date, training outcomes, promotion, retirement, etc. Also ...
3
votes
1
answer
37
views
Estimating the percentage of a specific factor being related to a condition
I am working on a medical research project where I am trying to find the associations between specific maternal factors and neonatal limb ischemia (NLI) to declare the maternal factors as "risk ...
- 207
2
votes
1
answer
38
views
Is it okay to leave a treatment group out of an analysis?
I am currently analyzing data from an RCT. The primary findings from the RCT have been published and we are now examining the data further. In brief, the RCT examined the effectiveness of a new ...
0
votes
0
answers
18
views
What statistical design and analysis is appropriate for this single-arm study?
I have research to perform but I am not very familiar with statistical design and the relevant analysis that I have to do, but here is the idea of what I want to do:
It is a single-arm study, I have a ...
- 1
0
votes
0
answers
42
views
SPSS Statistics HELP - Likert data
I have a cross sectional sample of around 300 respondents. The sample collects data on individuals living with a certain mental health disorder.
The variables are as follows:
Continuous dependent ...
1
vote
0
answers
17
views
Examples of statistical research affected by errors in package/library code
I can't figure out if this belongs in Cross Validated, Stack Overflow, or Academia, but I think this may be a good place to start:
Can anyone suggest examples of published prospective/retrospective/...
0
votes
0
answers
15
views
Dependent Variable in Factor Analysis. How to measure attitude toward application usage?
I am performing an analysis to evaluate the attitude of our marketers toward Apps. I have done the explanatory factor analysis (principal component)and I have received 4 factors (one of my factors is ...
1
vote
1
answer
201
views
McNemar test or Two proportion z-test?
Consider an experiment where a group of 200 people were asked before and after what is their preference, with possible outcomes being A and B. The research question is:
Is there a difference between ...
- 13
0
votes
0
answers
12
views
Comparing group means with uneven number of repeated measures
I have data on N different athletes each of which belongs to one of four groups. Each athlete was tested k_i times, i=1, ..., N. Unfortunately in general k_i != k_j for i != j. As far as I'm aware, ...
- 1
0
votes
0
answers
32
views
How to communicate comparing accuracy metrics between two different studies?
I'm working on a paper where we trained a statistical model and tested it on a dataset. We achieved some accuracy score.
There's another study that has a different accuracy score but in the same ...
0
votes
0
answers
36
views
What test can I use to compare experimental group 1 to placebo and to experimental group 2 but not compare experimental group 2 to the placebo group?
The aim of this study is to examine the difference between two different polyphenol supplements on exercise recovery (measurements taken at pre, immediately post, 1 week and 2 weeks post-)
The study ...
- 1
0
votes
0
answers
20
views
How to deal with redundant data within a sample when calculating the standard error?
In my research, I analyse pixel data from a 2D microscope image. Each pixel represents a unique spatial position. In my image, each pixel $i$ has a corresponding brightness value $x_i$. I assume the ...
- 1
1
vote
0
answers
18
views
Effect size and statistical significance correspondence
While I understand that it is possible to have a small effect size with statistical significance and a large effect size without statistical significance, is it ever possible to find this across two ...
- 11
1
vote
1
answer
22
views
Defining an independent variable?
I am conducting a small study where one class is taught similar (but not the same) concepts over two separate lessons.
The first lesson, the teacher will deliver the lesson to the class.
The second ...
- 13
0
votes
0
answers
77
views
Difference in difference with multiple treatments
I have a setting in which there are three disjoint groups as follows; (1) a group which does not experience any change, i.e., controls, (2) a group who receives treatment 1, and (3) and a group who ...
- 11
1
vote
0
answers
102
views
Ordinal Regression - Using a Continuous Attribute of 3 Categories as a Covariate as a Proxy for the Categories
I'm conducting an ordinal regression with SPSS 28. I have 3 subsets in my data (n=87 total, mostly equally distributed between the subsets) that represent 3 different scenarios where study ...
- 73
1
vote
2
answers
114
views
In a diff-in-diff or regression discontinuity research design, why is it important to describe why the counterfactual is a plausible one?
I've heard it mentioned that in difference-in-differences, regression discontinuity, or even in some other quasi-experimental research designs, that the counterfactual should be explained as a ...
- 111