Agree to Disagree? The Effect of Policy Alignment and Voice Pitch on Gendered Perceptions of Politicians

A hypothetical candidate's policy positions in a different voice pitch.

Participants will be randomly exposed to one of six audio recordings where a hypothetical candidate (male or female) discusses a gender-neutral policy issue. These recordings vary by the gender of the speaker, the pitch of the voice (low or high), and the stance on the policy (advocating or opposing):

T1: Female candidate advocating the same policy with a low-pitched voice.
T2: Female candidate opposing the policy with a low-pitched voice.
T3: Female candidate advocating the policy with a high-pitched voice.
T4: Female candidate opposing the policy with a high-pitched voice.
Control 1: Male candidate advocating with a low-pitched voice.
Control 2: Female candidate advocating a different policy with a low-pitched voice.

The script for these recordings is crafted in collaboration with a local politician and subsequently manipulated for pitch using vocal technology to ensure reality and external validity.

2024-05-31

2024-06-30

1) the likelihood of voting for the candidate; 2) the perceived potential of the candidate; and 3) the assessment of the candidate’s capability.

Participants are randomly selected from an opt-in online panel maintained by a Japanese survey firm, stratified by age (6 groups), region (8 regions), and sex (male and female). After indicating their agreement or disagreement with a gender-neutral policy issue, participants will listen to one of the six randomized recordings (see "Intervention"). They will then evaluate the candidate. Post-evaluation, agreement or disagreement with the candidate's policy stance is recorded along with demographic information.

Participants are randomly selected from an opt-in online panel maintained by a Japanese survey firm, stratified by age (6 groups; 18-29, 30-39, 40-49, 50-59, 60-69, 70 and above), region (8 regions), and sex (male and female).

After indicating their agreement or disagreement with a gender-neutral policy issue, participants will listen to one of the six randomized recordings (see "Intervention").

Policy agreements are measured using a 5-point scale: Agree, Somewhat Agree, Neither Agree nor Disagree, Somewhat Disagree, and Disagree. We then create a dummy variable that assigns one ("Agree") to respondents who either agree or somewhat agree, and zero ("Disagree") otherwise. This approach includes "Neither Agree nor Disagree" in the "Disagree" category, based on previous social psychology studies: it has been documented that Japanese respondents often select the midpoint on 5-point Likert scales. Especially concerning social policies, those who somewhat disagree often choose "Neither Agree nor Disagree" due to social desirability biases (Ogawa 2020).

They will then evaluate the candidate on three scales:

1) Likelihood of voting for the candidate.
2) Perceived potential of the candidate as a future leader.
3) Assessment of the candidate’s capability.

All three variables are measured on a 4-point scale ( “Very likely to vote” – “Very unlikely to vote” for the first variable; “Very likely to succeed as a future political leader” – “Very unlikely to succeed as a future political leader” for the second variable; “Very competent” – “Very incompetent” for the third variable).

Post-evaluation, agreement or disagreement with the candidate's policy stance is recorded along with demographic information (party affiliation, sex, education, age, height, self-reported voice pitch). We also track whether participants listened to the audio completely, skipped portions, or replayed the audio multiple times. We will examine whether this information moderates or mediates the results.

Citation:
Ogawa, K. (2020) "Does the Mid-point of Likert Scale Have an Impact on the Opinion on Social Policy?" Sociological Theory and Methods

Randomization will be carried out by the survey software Qualtrics.

Individual

No

2500 individuals (the treatment is not clustered)

2500 individuals

T1: 500
T2: 500
T3: 500
T4: 500
Control 1: 250
Control 2: 250