For the purpose of selecting the best one among the 4 design approaches, experimental video surveillance has taken. The dependent variable is the distance (meter) and the independent variables are lighting (3 levels) and design ( 4 levels). Perceptions on the distance have taken from the 100 respondents. For example, one pedestrian wearing a blue jacket (one design among 4) and walking on the straight road. A video was taken and prepared for respondent perception for showing on the computer screen. Each 5-meter distance from the video camera up to 200 m ( 40 sections) was shown and respondent gave their response as they detect or not. One respondent data like that: 5, 10, 15, 20, 25, 0, 35, 40, 45, 50, 55, 0, 0, 70, ...................................... Among the digits, 0 means unable to detect. I would like to select the best design approach based on the lighting. Additionally, How can I measure the effectiveness of 0?
When comparing sets of numerical data, a multiple group comparison test such as one-way analysis of variance (ANOVA) or Kruskal-Wallis test should be used first.
Dear Abdur Rahman ,
You may be able to download from
https://acutecaretesting.org/en/articles/statistical-analysis-in-method-comparison-studies-part-one
Regards,
Shafagat
Sounds like a job for a (repeated measurement) ANOVA to me. Please check whether the requirements of the ANOVA are fulfilled by your data.
When comparing sets of numerical data, a multiple group comparison test such as one-way analysis of variance (ANOVA) or Kruskal-Wallis test should be used first.
Read this article:
How to choose the right statistical test?
By : Barun K Nayak and Avijit Hazra
As few friends have advised you to use Anova, yes u can go ahead as there u will get interactions also along with main factors plots.
Secondly , since u r looking for optimizing your response based upon lightning so give priority to this factor in graphical study of response vs factors and response vs interactions.
Thors , since you r looking for effectiveness of `0` its mean other than `0` are one cluster so u can use two sample t test also. One cluste only having `0`, rest cluster all other sets of data.
This is as in `Anova`u will get there is a difference in responses which are statistically different but which one is best or no then u wont be able to come to know.
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis