Methods 911

Question 1

How do you translate research questions into directional and non-directional hypotheses?

Answer 1

Directional hypotheses predict a specific direction of effect (e.g., X will increase Y), while non-directional hypotheses only predict a relationship without specifying a direction (e.g., X affects Y)​(SOWO911_Week12_Fall2023…)​(SOWO911_Quantitative Da…).

Question 2

What are the implications of using directional versus non-directional hypotheses?

Answer 2

Directional hypotheses increase statistical power but require more certainty about the expected effect. Non-directional hypotheses are more conservative, providing flexibility when the effect’s direction is unclear​(SOWO911_Quantitative Da…).

Question 3

What is the NHST process?

Answer 3

NHST involves formulating the null and alternative hypotheses, selecting a significance level, conducting a test, and interpreting results based on whether the p-value falls below the alpha threshold​(SOWO911_Quantitative Da…)​(SOWO911_Week_12_Fall202…).

Question 4

What are the commonly used statistical significance levels in social science?

Answer 4

Common significance levels are 0.01, 0.05, and 0.10, corresponding to 99%, 95%, and 90% confidence levels, respectively​(SOWO911_Week_12_Fall202…).

Question 5

What are the consequences of using one-tailed vs. two-tailed tests?

Answer 5

One-tailed tests are more powerful for detecting effects in a specific direction but may miss effects in the opposite direction. Two-tailed tests are more flexible but require more data to achieve the same level of power​(SOWO_911_PPT_Slides_Fal…)

Question 6

For which research questions are T-tests and ANOVA appropriate?

Answer 6

T-tests are used to compare the means of two groups, while ANOVAs are used for comparing the means across three or more groups​(SOWO911_Week12_Fall2023…)​(SOWO911_Week_12_Fall202…).

Question 7

What are the assumptions for T-tests, ANOVA, and correlations?

Answer 7

Assumptions include normality, homogeneity of variance, and independent observations for T-tests and ANOVAs. Correlations assume linear relationships and continuous variables​(SOWO_911_PPT_Slides_Fal…)​(SOWO911_Quantitative Da…).

Question 8

What are the pros and cons of parametric vs. non-parametric tests?

Answer 8

Parametric tests are more powerful but assume normal distribution. Non-parametric tests are less powerful but are useful when assumptions of normality are violated​(SOWO911_Week12_Fall2023…)​(SOWO911_Week_12_Fall202…).

Question 9

What factors affect statistical power?

Answer 9

Statistical power increases with larger sample sizes, higher alpha levels, and stronger effect sizes. Power analysis helps determine the appropriate sample size needed to detect an effect​(SOWO911_Week_12_Fall202…)​(SOWO_911_PPT_Slides_Fal…).

Question 10

What types of correlation tests exist, and when are they used?

Answer 10

Pearson is used for interval/ratio variables. Spearman is for ordinal variables. Polychoric and biserial correlations are used for combinations of ordinal, dichotomous, and continuous data​(SOWO_911_PPT_Slides_Fal…).

Question 11

How do you translate research questions into directional and non-directional hypotheses?

Answer 11

A directional hypothesis predicts a specific direction of effect (e.g., “X will increase Y”), while a non-directional hypothesis predicts a relationship without specifying the direction (e.g., “X will affect Y”).​(SOWO911_Week12_Fall2023…)

Question 12

What is the difference between a research hypothesis and a statistical hypothesis?

Answer 12

Question 13

What are the implications of using directional versus non-directional hypotheses?

Answer 13

Directional hypotheses are more powerful because they focus on one direction of an effect, but they risk missing an effect in the opposite direction. Non-directional hypotheses are more conservative but provide greater flexibility, testing for effects in both directions​(SOWO911_Week12_Fall2023…).

Question 14

What are the steps in the Null Hypothesis Significance Testing (NHST) process?

Answer 14

Question 15

What are common statistical significance levels and critical values in social science?

Answer 15

Question 16

What are the consequences of using one-tailed versus two-tailed tests?

Answer 16

A one-tailed test is more powerful if the effect is expected in a specific direction but risks missing effects in the opposite direction. A two-tailed test is more conservative, accounting for effects in both directions but requiring more data to achieve significance​(SOWO911_Week12_Fall2023…)​(SOWO911_Week_12_Fall202…).

Question 17

When should you use a one-sample, independent sample, or paired sample t-test?

Answer 17

Question 18

What are between-subjects and within-subjects ANOVAs?

Answer 18

Question 19

What are the different types of Chi-square tests and when are they used?

Answer 19

Question 20

What are Pearson, Spearman, and other correlation tests used for?

Answer 20

Question 21

What are the assumptions for t-tests, ANOVAs, and correlation tests?

Answer 21

Question 22

How can you justify ignoring minor violations of assumptions in statistical tests?

Answer 22

Minor violations can be ignored if the test is robust to the violation, especially in large samples. Robust tests like ANOVA and t-tests can handle slight deviations from normality and homogeneity​(SOWO_911_PPT_Slides_Fal…)​(SOWO911_Week_12_Fall202…).

Question 23

What are the pros and cons of parametric versus non-parametric tests?

Answer 23

Question 24

What is an example of a research question and hypothesis related to maternal health and climate change from a social work perspective?

Answer 24

Question 25

Provide examples of a directional and non-directional hypothesis related to climate change and maternal health in underserved communities.

Answer 25

Directional Hypothesis: Women in communities disproportionately affected by flooding will have higher rates of maternal mental health disorders. Non-Directional Hypothesis: Extreme weather events will impact maternal mental health outcomes, but the direction of the effect is unknown.

Question 26

What statistical test could you use to evaluate the effectiveness of a social work intervention designed to support maternal health in climate-vulnerable communities?

Answer 26

A paired sample t-test could be used to compare stress levels before and after a social work intervention aimed at providing support to pregnant women during extreme weather events.

Question 27

How can a chi-square test be applied to assess the relationship between social work interventions and maternal health outcomes in the context of climate change?

Answer 27

A chi-square test could be used to evaluate whether women who receive social work support during extreme weather (e.g., access to safe housing) experience lower rates of preterm birth compared to those who don’t.

Question 28

What type of correlation test would you use to assess the relationship between access to social services and maternal health outcomes in the face of climate-related stress?

Answer 28

A Spearman correlation could be used to examine the relationship between access to community support services (ordinal variable) and prenatal stress levels (continuous variable) in climate-affected regions.

Question 29

What assumptions should you check when analyzing the impact of social interventions on maternal health in climate-affected communities?

Answer 29

A: You should ensure: Normality of maternal health outcome data (e.g., mental health scores). Homogeneity of variance across communities with and without social work interventions. Independent observations of participants in different intervention groups.

Question 30

When might a social worker use a one-tailed versus a two-tailed test to evaluate the impact of social work interventions on maternal health outcomes?

Answer 30

A one-tailed test could be used if the intervention is expected to reduce maternal health complications in areas affected by climate change, while a two-tailed test would be appropriate if you’re unsure whether the intervention will increase or decrease complications

Question 31

How can social workers decide between parametric and non-parametric tests when analyzing climate change's impact on maternal health?

Answer 31

Parametric tests (e.g., t-tests) are more powerful when data assumptions (normality, homogeneity) are met. Non-parametric tests (e.g., Mann-Whitney U) are useful when working with skewed data or small, vulnerable populations where assumptions may not hold.

Question 32

Answer 32

Question 33

Answer 33

Question 34

Answer 34

Question 35

Paired Samples t-test:

Answer 35

Question 36

3. ANOVA (Analysis of Variance) Equation:

Answer 36

Question 37

Chi-Square Test:

Answer 37

Question 38

Answer 38

Question 39

Why these equations matter

Answer 39

Question 40

When do you use a one-sample t-test?

Answer 40

Use a one-sample t-test when you want to compare the mean of a sample to a known population mean. Example: Comparing the average birth weight of infants in a region to the national average birth weight.

Question 41

Q: When do you use an independent samples t-test?

Answer 41

A: Use an independent samples t-test to compare the means of two independent groups. Example: Comparing the average stress levels of pregnant women in high-pollution vs. low-pollution areas.

Question 42

When do you use a one-way ANOVA?

Answer 42

Use a one-way ANOVA when you want to compare the means of three or more independent groups based on one factor (e.g., exposure to different levels of air pollution). Example: Comparing birth weights across three regions with different levels of air pollution.

Question 43

When do you use a paired samples t-test?

Answer 43

Use a paired samples t-test when comparing two measurements taken from the same group of individuals at different times (before/after treatment). Example: Measuring maternal stress levels before and after a climate-related intervention.

Question 44

When do you use a two-way ANOVA?

Answer 44

Use a two-way ANOVA when you want to assess the effect of two independent factors on a dependent variable and see if there is an interaction between them. Example: Comparing maternal health outcomes based on both region (urban/rural) and climate exposure (high/low).

Question 45

When do you use a chi-square test?

Answer 45

A: Use a chi-square test when you want to examine the relationship between two categorical variables. Example: Assessing whether there is an association between extreme heat exposure (yes/no) and preterm birth (yes/no).

Question 46

When do you use Pearson correlation?

Answer 46

A: Use Pearson correlation when you want to measure the strength and direction of a linear relationship between two continuous variables. Example: Assessing the relationship between average temperature and maternal anemia levels.

Question 47

When do you use Spearman correlation?

Answer 47

A: Use Spearman correlation when you want to assess the relationship between two ordinal variables or when the data does not meet the assumptions of normality for Pearson correlation. Example: Assessing the relationship between ranked access to healthcare and maternal health outcomes.

Question 48

How do you know when to use a paired sample t-test versus an independent sample t-test?

Answer 48

Use a paired sample t-test if you’re measuring the same group at two different times or under two different conditions. Use an independent sample t-test if you’re comparing two different groups of individuals. Example: Paired sample: Comparing maternal stress before and after a social intervention. Independent sample: Comparing maternal stress between two different populations.

Question 49

When do you use a one-tailed test versus a two-tailed test?

Answer 49

A: Use a one-tailed test when you expect an effect in a specific direction (e.g., greater stress in hotter climates). Use a two-tailed test if you're open to effects in either direction (e.g., climate could increase or decrease stress). Example: One-tailed: Testing if high temperatures increase preterm birth. Two-tailed: Testing if temperatures affect preterm birth rates without assuming a specific direction.

Question 50

How do you know whether to use an ANOVA or a t-test?

Answer 50

A: Use a t-test when comparing the means of two groups. Use ANOVA when comparing the means of three or more groups. Example: t-test: Comparing maternal health in rural vs. urban settings. ANOVA: Comparing maternal health across rural, suburban, and urban settings.

Question 51

Q: What is a nominal scale?

Answer 51

A: A nominal scale is a categorical scale where values are used as labels without any quantitative value or order. The categories are distinct and mutually exclusive. Example: Gender (male, female), marital status (married, single).

Question 52

: What is an ordinal scale?

Answer 52

A: An ordinal scale is a categorical scale where values have a meaningful order, but the intervals between values are not equal or known. Example: Education level (high school, bachelor’s, master’s), pain scale (mild, moderate, severe).

Question 53

Q: What is an interval scale?

Answer 53

A: An interval scale has ordered values with known, equal intervals between them, but no true zero point (meaning zero does not indicate the absence of the variable). Example: Temperature in Celsius or Fahrenheit, IQ scores.

Question 54

What is a ratio scale?

Answer 54

A: A ratio scale has ordered values, equal intervals, and a true zero point (where zero means none of the variable exists). Example: Height, weight, income.