exam 2 study pt 2

quasi ppt and survey ppt

describe a quasi experimental design

•close to experimental design, but lack random assignment and/or a comparison group

–Often having nonequivalent groups

–Degree of control is decreased

describe a one group pretest post test design

• does not have a control group, so One group is measured before and after treatment

• IV = time (pretest vs posttest)

• Weak internal validity and external validity due to lack of control group

• Used when control group is not feasible or prior evidence exists

---

one way repeated measure design

You measure the same outcome many times over time, and you look at how it changes when (or if) a treatment is introduced.

• IV = time

• No randomization or control group

• Useful when disease progression is predictable

• Weak internal validity (cannot strongly infer causation)

describe a time series design

•Multiple measurements before and after intervention to examine patterns of behaviors

•Most effective when serial data can be collected evenly (time) throughout the study

–Avoids confounding variables

–Most threats to internal validity are controlled for

–External validity is limited

describe the non equivalent pretest posttest control group design

•Similar to pretest-posttest experimental design, except subjects are not randomly assigned

•Can be designed with one treatment group and one control group or with multiple of each

•Limited due to lack of randomization

historical controls

instead of comparing two groups at the same time, you compare a new treatment group to a past group of patients who already received the old treatment.

The control group is NOT recruited at the same time—it comes from earlier data.

• Used when researchers expect a new treatment to be superior or when prospective controls are not feasible

• Advantage: efficient, uses existing data

• Disadvantage: low internal validity due to differences across time; findings may not replicate in RCTs

non equivalent post test only control group design

•Should only be used for exploratory research- cannot prove cause effect

•Can include any number of treatment levels, with or without a control group

Internal validity is severely threatened by selection bias and attrition

ex: A researcher studies a rehab program:

• Group 1: patients from Clinic A (gets new therapy)

• Group 2: patients from Clinic B (standard therapy)

They measure outcomes only at the end and compare groups.

define interview, survey, and questionnaires

survey- composed of questions that can be given to a group( intended for generalization/description of a group)

interview- asks respondents questions and record answers for analysis ( typically in respondants natural setting

questionannaires- structured surveys that are self administered

what are the advantages and disadvatages of interviews

oAdvantages

oIn-depth analysis

oCan probe responses

oCan directly observe respondents’ reactions

o Disadvantages

oCost and time

oNeed for personnel

oScheduling

oLack of anonymity

what are the questionnaire advantages and disadvantages

oAdvantages

oMore efficient than interviews

oData can be gathered from a large sample (geographically)

oShort period of time to gather data

oRespondents can take time to choose answer

oProvide anonymity

oUseful to examine phenomena that cannot be observed

oDisadvantages

Potential misunderstanding of question

what is the distribution, return rate, self report for questionnares

Distribution

• Mail

• Online / survey platforms

Response rates

• 60–80% = excellent

• 30–60% = common

• Low return → incomplete data often discarded

Self-report issues

• Recall bias (forgetting past events)

• Social desirability bias

• Inaccuracy possible (but still widely used)

what is the design of surveys process

1. research question

2. hypothesis

3. questionnaire outline

4. review of existing instruments

5. designing the instrument ( logical order)

6. preliminary drafts (panel feedback)

7. pilot testing and revisions

8. selecting a sample

what is an open ended question and close ended question

• Respond in own words

• rich detail

• − hard to code/analyze

Closed-ended

• Fixed answer choices

• easy to code, consistent

• − limits responses, may bias

what are the two common formats of close ended questions

brackets of circling answer

what are the formats of close ended questions

• Dichotomous (Yes/No)

• Range of responses (ex: pain scale)

• Grid/checklist

• Ranking order (1 = most important)

• Branching (skip irrelevant questions) ex: picture

what are the survey question writing rules

• Keep simple language

• Avoid jargon/idioms

• Avoid double-barreled questions (2 questions in 1)

• Be clear + unambiguous

• Include time frame for frequency questions

• Be careful with sensitive questions

  • allow refusal option

  • reduce discomfort

what is the purpose of scales

To distinguish between people who have different intensities of the measured characteristic

what are the types of scales

Categorical

• Groups (gender, diagnosis)

• Frequencies/percentages

Continuous

• Measured on continuum (age, BP)

Summative scales

• Total score from multiple items

• All items equally weighted

Cumulative scales

• Items build in intensity (e.g., depression severity)

what is a linkert scale

• Most common summative scale to assess attitudes/values

• “Strongly agree → strongly disagree”

• Produces total summed score

• Requires item analysis/validation

what is a semantic differential scale

• Measures attitude using opposite adjectives

  • e.g., good – bad, strong – weak

• Usually 7-point scale

• Not agree/disagree (different from Likert)

analysis of survey data process

1. code responses into computer

2. Responses to closed-ended questions are coded

3. analysis can look like » Descriptive statistic, Relationships between variables, Totals from scales

what is the relation between informed consent an surveys

• Must go through IRB (review board)

• Surveys often exempt or expedited

• Still required even without face-to-face contact